Semantic Segmentation with Model Garden

View on TensorFlow.org Run in Google Colab View on GitHub Download notebook

This tutorial trains a DeepLabV3 with Mobilenet V2 as backbone model from the TensorFlow Model Garden package (tensorflow-models).

Model Garden contains a collection of state-of-the-art models, implemented with TensorFlow's high-level APIs. The implementations demonstrate the best practices for modeling, letting users to take full advantage of TensorFlow for their research and product development.

Dataset: Oxford-IIIT Pets

  • The Oxford-IIIT pet dataset is a 37 category pet image dataset with roughly 200 images for each class. The images have large variations in scale, pose and lighting. All images have an associated ground truth annotation of breed.

This tutorial demonstrates how to:

  1. Use models from the TensorFlow Models package.
  2. Train/Fine-tune a pre-built DeepLabV3 with mobilenet as backbone for Semantic Segmentation.
  3. Export the trained/tuned DeepLabV3 model

Install necessary dependencies

pip install -U -q "tf-models-official"

Import required libraries

import os
import pprint
import numpy as np
import matplotlib.pyplot as plt

from IPython import display

import tensorflow as tf
import tensorflow_datasets as tfds


import orbit
import tensorflow_models as tfm
from official.vision.data import tfrecord_lib
from official.vision.utils import summary_manager
from official.vision.serving import export_saved_model_lib
from official.vision.utils.object_detection import visualization_utils

pp = pprint.PrettyPrinter(indent=4) # Set Pretty Print Indentation
print(tf.__version__) # Check the version of tensorflow used

%matplotlib inline

2024-02-02 12:12:13.799558: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2024-02-02 12:12:13.799625: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2024-02-02 12:12:13.801330: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2.15.0

Custom dataset preparation for semantic segmentation

Models in Official repository (of model-garden) require models in a TFRecords dataformat.

Please check this resource to learn more about TFRecords data format.

Oxford_IIIT_pet:3 dataset is taken from Tensorflow Datasets

(train_ds, val_ds, test_ds), info = tfds.load(
    'oxford_iiit_pet:3.*.*',
    split=['train+test[:50%]', 'test[50%:80%]', 'test[80%:100%]'],
    with_info=True)
info

tfds.core.DatasetInfo(
    name='oxford_iiit_pet',
    full_name='oxford_iiit_pet/3.2.0',
    description="""
    The Oxford-IIIT pet dataset is a 37 category pet image dataset with roughly 200
    images for each class. The images have large variations in scale, pose and
    lighting. All images have an associated ground truth annotation of breed.
    """,
    homepage='http://www.robots.ox.ac.uk/~vgg/data/pets/',
    data_dir='gs://tensorflow-datasets/datasets/oxford_iiit_pet/3.2.0',
    file_format=tfrecord,
    download_size=773.52 MiB,
    dataset_size=774.69 MiB,
    features=FeaturesDict({
        'file_name': Text(shape=(), dtype=string),
        'image': Image(shape=(None, None, 3), dtype=uint8),
        'label': ClassLabel(shape=(), dtype=int64, num_classes=37),
        'segmentation_mask': Image(shape=(None, None, 1), dtype=uint8),
        'species': ClassLabel(shape=(), dtype=int64, num_classes=2),
    }),
    supervised_keys=('image', 'label'),
    disable_shuffling=False,
    splits={
        'test': <SplitInfo num_examples=3669, num_shards=4>,
        'train': <SplitInfo num_examples=3680, num_shards=4>,
    },
    citation="""@InProceedings{parkhi12a,
      author       = "Parkhi, O. M. and Vedaldi, A. and Zisserman, A. and Jawahar, C.~V.",
      title        = "Cats and Dogs",
      booktitle    = "IEEE Conference on Computer Vision and Pattern Recognition",
      year         = "2012",
    }""",
)

Helper function to encode dataset as tfrecords

def process_record(record):
  keys_to_features = {
  'image/encoded': tfrecord_lib.convert_to_feature(
      tf.io.encode_jpeg(record['image']).numpy()),
  'image/height': tfrecord_lib.convert_to_feature(record['image'].shape[0]),
  'image/width': tfrecord_lib.convert_to_feature(record['image'].shape[1]),
  'image/segmentation/class/encoded':tfrecord_lib.convert_to_feature(
      tf.io.encode_png(record['segmentation_mask'] - 1).numpy())
  }
  example = tf.train.Example(
      features=tf.train.Features(feature=keys_to_features))
  return example

Write TFRecords to a folder

output_dir = './oxford_iiit_pet_tfrecords/'
LOG_EVERY = 100
if not os.path.exists(output_dir):
  os.mkdir(output_dir)

def write_tfrecords(dataset, output_path, num_shards=1):
  writers = [
        tf.io.TFRecordWriter(
            output_path + '-%05d-of-%05d.tfrecord' % (i, num_shards))
        for i in range(num_shards)
    ]
  for idx, record in enumerate(dataset):
    if idx % LOG_EVERY == 0:
      print('On image %d', idx)
    tf_example = process_record(record)
    writers[idx % num_shards].write(tf_example.SerializeToString())

Write training data as TFRecords

output_train_tfrecs = output_dir + 'train'
write_tfrecords(train_ds, output_train_tfrecs, num_shards=10)

On image %d 0
On image %d 100
On image %d 200
On image %d 300
On image %d 400
On image %d 500
On image %d 600
On image %d 700
On image %d 800
Corrupt JPEG data: 240 extraneous bytes before marker 0xd9
Corrupt JPEG data: premature end of data segment
On image %d 900
On image %d 1000
On image %d 1100
On image %d 1200
On image %d 1300
On image %d 1400
On image %d 1500
On image %d 1600
On image %d 1700
On image %d 1800
On image %d 1900
On image %d 2000
On image %d 2100
On image %d 2200
On image %d 2300
On image %d 2400
On image %d 2500
On image %d 2600
On image %d 2700
On image %d 2800
On image %d 2900
On image %d 3000
On image %d 3100
On image %d 3200
On image %d 3300
On image %d 3400
On image %d 3500
On image %d 3600
On image %d 3700
On image %d 3800
On image %d 3900
On image %d 4000
On image %d 4100
On image %d 4200
On image %d 4300
On image %d 4400
On image %d 4500
On image %d 4600
On image %d 4700
On image %d 4800
On image %d 4900
On image %d 5000
On image %d 5100
On image %d 5200
On image %d 5300
On image %d 5400
On image %d 5500

Write validation data as TFRecords

output_val_tfrecs = output_dir + 'val'
write_tfrecords(val_ds, output_val_tfrecs, num_shards=5)

On image %d 0
On image %d 100
On image %d 200
On image %d 300
On image %d 400
On image %d 500
On image %d 600
On image %d 700
On image %d 800
On image %d 900
On image %d 1000
On image %d 1100

Write test data as TFRecords

output_test_tfrecs = output_dir + 'test'
write_tfrecords(test_ds, output_test_tfrecs, num_shards=5)

On image %d 0
On image %d 100
On image %d 200
On image %d 300
On image %d 400
On image %d 500
On image %d 600
On image %d 700

Configure the DeepLabV3 Mobilenet model for custom dataset

train_data_tfrecords = './oxford_iiit_pet_tfrecords/train*'
val_data_tfrecords = './oxford_iiit_pet_tfrecords/val*'
test_data_tfrecords = './oxford_iiit_pet_tfrecords/test*'
trained_model = './trained_model/'
export_dir = './exported_model/'

In Model Garden, the collections of parameters that define a model are called configs. Model Garden can create a config based on a known set of parameters via a factory.

Use the mnv2_deeplabv3_pascal experiment configuration, as defined by tfm.vision.configs.semantic_segmentation.mnv2_deeplabv3_pascal.

Please find all the registered experiements here

The configuration defines an experiment to train a DeepLabV3 model with MobilenetV2 as backbone and ASPP as decoder.

There are also other alternative experiments available such as

  • seg_deeplabv3_pascal
  • seg_deeplabv3plus_pascal
  • seg_resnetfpn_pascal
  • mnv2_deeplabv3plus_cityscapes

and more. One can switch to them by changing the experiment name argument to the get_exp_config function.

exp_config = tfm.core.exp_factory.get_exp_config('mnv2_deeplabv3_pascal')

model_ckpt_path = './model_ckpt/'
if not os.path.exists(model_ckpt_path):
  os.mkdir(model_ckpt_path)

!gsutil cp gs://tf_model_garden/cloud/vision-2.0/deeplab/deeplabv3_mobilenetv2_coco/best_ckpt-63.data-00000-of-00001 './model_ckpt/'
!gsutil cp gs://tf_model_garden/cloud/vision-2.0/deeplab/deeplabv3_mobilenetv2_coco/best_ckpt-63.index './model_ckpt/'

Copying gs://tf_model_garden/cloud/vision-2.0/deeplab/deeplabv3_mobilenetv2_coco/best_ckpt-63.data-00000-of-00001...

Operation completed over 1 objects/28.2 MiB.                                     
Copying gs://tf_model_garden/cloud/vision-2.0/deeplab/deeplabv3_mobilenetv2_coco/best_ckpt-63.index...

Operation completed over 1 objects/12.5 KiB.

Adjust the model and dataset configurations so that it works with custom dataset.

num_classes = 3
WIDTH, HEIGHT = 128, 128
input_size = [HEIGHT, WIDTH, 3]
BATCH_SIZE = 16

# Backbone Config
exp_config.task.init_checkpoint = model_ckpt_path + 'best_ckpt-63'
exp_config.task.freeze_backbone = True

# Model Config
exp_config.task.model.num_classes = num_classes
exp_config.task.model.input_size = input_size

# Training Data Config
exp_config.task.train_data.aug_scale_min = 1.0
exp_config.task.train_data.aug_scale_max = 1.0
exp_config.task.train_data.input_path = train_data_tfrecords
exp_config.task.train_data.global_batch_size = BATCH_SIZE
exp_config.task.train_data.dtype = 'float32'
exp_config.task.train_data.output_size = [HEIGHT, WIDTH]
exp_config.task.train_data.preserve_aspect_ratio = False
exp_config.task.train_data.seed = 21 # Reproducable Training Data

# Validation Data Config
exp_config.task.validation_data.input_path = val_data_tfrecords
exp_config.task.validation_data.global_batch_size = BATCH_SIZE
exp_config.task.validation_data.dtype = 'float32'
exp_config.task.validation_data.output_size = [HEIGHT, WIDTH]
exp_config.task.validation_data.preserve_aspect_ratio = False
exp_config.task.validation_data.groundtruth_padded_size = [HEIGHT, WIDTH]
exp_config.task.validation_data.seed = 21 # Reproducable Validation Data
exp_config.task.validation_data.resize_eval_groundtruth = True # To enable validation loss

Adjust the trainer configuration.

logical_device_names = [logical_device.name
                        for logical_device in tf.config.list_logical_devices()]

if 'GPU' in ''.join(logical_device_names):
  print('This may be broken in Colab.')
  device = 'GPU'
elif 'TPU' in ''.join(logical_device_names):
  print('This may be broken in Colab.')
  device = 'TPU'
else:
  print('Running on CPU is slow, so only train for a few steps.')
  device = 'CPU'


train_steps = 2000
exp_config.trainer.steps_per_loop = int(train_ds.__len__().numpy() // BATCH_SIZE)

exp_config.trainer.summary_interval = exp_config.trainer.steps_per_loop # steps_per_loop = num_of_validation_examples // eval_batch_size
exp_config.trainer.checkpoint_interval = exp_config.trainer.steps_per_loop
exp_config.trainer.validation_interval = exp_config.trainer.steps_per_loop
exp_config.trainer.validation_steps =  int(train_ds.__len__().numpy() // BATCH_SIZE) # validation_steps = num_of_validation_examples // eval_batch_size
exp_config.trainer.train_steps = train_steps
exp_config.trainer.optimizer_config.warmup.linear.warmup_steps = exp_config.trainer.steps_per_loop
exp_config.trainer.optimizer_config.learning_rate.type = 'cosine'
exp_config.trainer.optimizer_config.learning_rate.cosine.decay_steps = train_steps
exp_config.trainer.optimizer_config.learning_rate.cosine.initial_learning_rate = 0.1
exp_config.trainer.optimizer_config.warmup.linear.warmup_learning_rate = 0.05

This may be broken in Colab.

pp.pprint(exp_config.as_dict())
display.Javascript('google.colab.output.setIframeHeight("500px");')

{   'runtime': {   'all_reduce_alg': None,
                   'batchnorm_spatial_persistent': False,
                   'dataset_num_private_threads': None,
                   'default_shard_dim': -1,
                   'distribution_strategy': 'mirrored',
                   'enable_xla': False,
                   'gpu_thread_mode': None,
                   'loss_scale': None,
                   'mixed_precision_dtype': None,
                   'num_cores_per_replica': 1,
                   'num_gpus': 0,
                   'num_packs': 1,
                   'per_gpu_thread_count': 0,
                   'run_eagerly': False,
                   'task_index': -1,
                   'tpu': None,
                   'tpu_enable_xla_dynamic_padder': None,
                   'use_tpu_mp_strategy': False,
                   'worker_hosts': None},
    'task': {   'allow_image_summary': True,
                'differential_privacy_config': None,
                'eval_input_partition_dims': [],
                'evaluation': {   'report_per_class_iou': True,
                                  'report_train_mean_iou': True},
                'export_config': {'rescale_output': False},
                'freeze_backbone': True,
                'init_checkpoint': './model_ckpt/best_ckpt-63',
                'init_checkpoint_modules': ['backbone', 'decoder'],
                'losses': {   'class_weights': [],
                              'gt_is_matting_map': False,
                              'ignore_label': 255,
                              'l2_weight_decay': 4e-05,
                              'label_smoothing': 0.0,
                              'loss_weight': 1.0,
                              'mask_scoring_weight': 1.0,
                              'top_k_percent_pixels': 1.0,
                              'use_binary_cross_entropy': False,
                              'use_groundtruth_dimension': True},
                'model': {   'backbone': {   'mobilenet': {   'filter_size_scale': 1.0,
                                                              'model_id': 'MobileNetV2',
                                                              'output_intermediate_endpoints': False,
                                                              'output_stride': 16,
                                                              'stochastic_depth_drop_rate': 0.0},
                                             'type': 'mobilenet'},
                             'decoder': {   'aspp': {   'dilation_rates': [],
                                                        'dropout_rate': 0.0,
                                                        'level': 4,
                                                        'num_filters': 256,
                                                        'output_tensor': False,
                                                        'pool_kernel_size': [],
                                                        'spp_layer_version': 'v1',
                                                        'use_depthwise_convolution': False},
                                            'type': 'aspp'},
                             'head': {   'decoder_max_level': None,
                                         'decoder_min_level': None,
                                         'feature_fusion': None,
                                         'level': 4,
                                         'logit_activation': None,
                                         'low_level': 2,
                                         'low_level_num_filters': 48,
                                         'num_convs': 0,
                                         'num_filters': 256,
                                         'prediction_kernel_size': 1,
                                         'upsample_factor': 1,
                                         'use_depthwise_convolution': False},
                             'input_size': [128, 128, 3],
                             'mask_scoring_head': None,
                             'max_level': 6,
                             'min_level': 3,
                             'norm_activation': {   'activation': 'relu',
                                                    'norm_epsilon': 0.001,
                                                    'norm_momentum': 0.99,
                                                    'use_sync_bn': True},
                             'num_classes': 3},
                'name': None,
                'train_data': {   'additional_dense_features': [],
                                  'apply_tf_data_service_before_batching': False,
                                  'aug_policy': None,
                                  'aug_rand_hflip': True,
                                  'aug_scale_max': 1.0,
                                  'aug_scale_min': 1.0,
                                  'autotune_algorithm': None,
                                  'block_length': 1,
                                  'cache': False,
                                  'crop_size': [],
                                  'cycle_length': 10,
                                  'decoder': {   'simple_decoder': {   'attribute_names': [   ],
                                                                       'mask_binarize_threshold': None,
                                                                       'regenerate_source_id': False},
                                                 'type': 'simple_decoder'},
                                  'deterministic': None,
                                  'drop_remainder': True,
                                  'dtype': 'float32',
                                  'enable_shared_tf_data_service_between_parallel_trainers': False,
                                  'enable_tf_data_service': False,
                                  'file_type': 'tfrecord',
                                  'global_batch_size': 16,
                                  'groundtruth_padded_size': [],
                                  'image_feature': {   'feature_name': 'image/encoded',
                                                       'mean': (   123.675,
                                                                   116.28,
                                                                   103.53),
                                                       'num_channels': 3,
                                                       'stddev': (   58.395,
                                                                     57.120000000000005,
                                                                     57.375)},
                                  'input_path': './oxford_iiit_pet_tfrecords/train*',
                                  'is_training': True,
                                  'output_size': [128, 128],
                                  'prefetch_buffer_size': None,
                                  'preserve_aspect_ratio': False,
                                  'resize_eval_groundtruth': True,
                                  'seed': 21,
                                  'sharding': True,
                                  'shuffle_buffer_size': 1000,
                                  'tf_data_service_address': None,
                                  'tf_data_service_job_name': None,
                                  'tfds_as_supervised': False,
                                  'tfds_data_dir': '',
                                  'tfds_name': '',
                                  'tfds_skip_decoding_feature': '',
                                  'tfds_split': '',
                                  'trainer_id': None,
                                  'weights': None},
                'train_input_partition_dims': [],
                'validation_data': {   'additional_dense_features': [],
                                       'apply_tf_data_service_before_batching': False,
                                       'aug_policy': None,
                                       'aug_rand_hflip': True,
                                       'aug_scale_max': 1.0,
                                       'aug_scale_min': 1.0,
                                       'autotune_algorithm': None,
                                       'block_length': 1,
                                       'cache': False,
                                       'crop_size': [],
                                       'cycle_length': 10,
                                       'decoder': {   'simple_decoder': {   'attribute_names': [   ],
                                                                            'mask_binarize_threshold': None,
                                                                            'regenerate_source_id': False},
                                                      'type': 'simple_decoder'},
                                       'deterministic': None,
                                       'drop_remainder': False,
                                       'dtype': 'float32',
                                       'enable_shared_tf_data_service_between_parallel_trainers': False,
                                       'enable_tf_data_service': False,
                                       'file_type': 'tfrecord',
                                       'global_batch_size': 16,
                                       'groundtruth_padded_size': [128, 128],
                                       'image_feature': {   'feature_name': 'image/encoded',
                                                            'mean': (   123.675,
                                                                        116.28,
                                                                        103.53),
                                                            'num_channels': 3,
                                                            'stddev': (   58.395,
                                                                          57.120000000000005,
                                                                          57.375)},
                                       'input_path': './oxford_iiit_pet_tfrecords/val*',
                                       'is_training': False,
                                       'output_size': [128, 128],
                                       'prefetch_buffer_size': None,
                                       'preserve_aspect_ratio': False,
                                       'resize_eval_groundtruth': True,
                                       'seed': 21,
                                       'sharding': True,
                                       'shuffle_buffer_size': 1000,
                                       'tf_data_service_address': None,
                                       'tf_data_service_job_name': None,
                                       'tfds_as_supervised': False,
                                       'tfds_data_dir': '',
                                       'tfds_name': '',
                                       'tfds_skip_decoding_feature': '',
                                       'tfds_split': '',
                                       'trainer_id': None,
                                       'weights': None} },
    'trainer': {   'allow_tpu_summary': False,
                   'best_checkpoint_eval_metric': 'mean_iou',
                   'best_checkpoint_export_subdir': 'best_ckpt',
                   'best_checkpoint_metric_comp': 'higher',
                   'checkpoint_interval': 344,
                   'continuous_eval_timeout': 3600,
                   'eval_tf_function': True,
                   'eval_tf_while_loop': False,
                   'loss_upper_bound': 1000000.0,
                   'max_to_keep': 5,
                   'optimizer_config': {   'ema': None,
                                           'learning_rate': {   'cosine': {   'alpha': 0.0,
                                                                              'decay_steps': 2000,
                                                                              'initial_learning_rate': 0.1,
                                                                              'name': 'CosineDecay',
                                                                              'offset': 0},
                                                                'type': 'cosine'},
                                           'optimizer': {   'sgd': {   'clipnorm': None,
                                                                       'clipvalue': None,
                                                                       'decay': 0.0,
                                                                       'global_clipnorm': None,
                                                                       'momentum': 0.9,
                                                                       'name': 'SGD',
                                                                       'nesterov': False},
                                                            'type': 'sgd'},
                                           'warmup': {   'linear': {   'name': 'linear',
                                                                       'warmup_learning_rate': 0.05,
                                                                       'warmup_steps': 344},
                                                         'type': 'linear'} },
                   'preemption_on_demand_checkpoint': True,
                   'recovery_begin_steps': 0,
                   'recovery_max_trials': 0,
                   'steps_per_loop': 344,
                   'summary_interval': 344,
                   'train_steps': 2000,
                   'train_tf_function': True,
                   'train_tf_while_loop': True,
                   'validation_interval': 344,
                   'validation_steps': 344,
                   'validation_summary_subdir': 'validation'} }
<IPython.core.display.Javascript object>

Set up the distribution strategy.

# Setting up the Strategy
if exp_config.runtime.mixed_precision_dtype == tf.float16:
    tf.keras.mixed_precision.set_global_policy('mixed_float16')

if 'GPU' in ''.join(logical_device_names):
  distribution_strategy = tf.distribute.MirroredStrategy()
elif 'TPU' in ''.join(logical_device_names):
  tf.tpu.experimental.initialize_tpu_system()
  tpu = tf.distribute.cluster_resolver.TPUClusterResolver(tpu='/device:TPU_SYSTEM:0')
  distribution_strategy = tf.distribute.experimental.TPUStrategy(tpu)
else:
  print('Warning: this will be really slow.')
  distribution_strategy = tf.distribute.OneDeviceStrategy(logical_device_names[0])

print("Done")

INFO:tensorflow:Using MirroredStrategy with devices ('/job:localhost/replica:0/task:0/device:GPU:0', '/job:localhost/replica:0/task:0/device:GPU:1', '/job:localhost/replica:0/task:0/device:GPU:2', '/job:localhost/replica:0/task:0/device:GPU:3')
INFO:tensorflow:Using MirroredStrategy with devices ('/job:localhost/replica:0/task:0/device:GPU:0', '/job:localhost/replica:0/task:0/device:GPU:1', '/job:localhost/replica:0/task:0/device:GPU:2', '/job:localhost/replica:0/task:0/device:GPU:3')
Done

Create the Task object (tfm.core.base_task.Task) from the config_definitions.TaskConfig.

The Task object has all the methods necessary for building the dataset, building the model, and running training & evaluation. These methods are driven by tfm.core.train_lib.run_experiment.

model_dir = './trained_model/'

with distribution_strategy.scope():
  task = tfm.core.task_factory.get_task(exp_config.task, logging_dir=model_dir)

Visualize a batch of the data.

for images, masks in task.build_inputs(exp_config.task.train_data).take(1):
  print()
  print(f'images.shape: {str(images.shape):16}  images.dtype: {images.dtype!r}')
  print(f'masks.shape: {str(masks["masks"].shape):16} images.dtype: {masks["masks"].dtype!r}')

images.shape: (16, 128, 128, 3)  images.dtype: tf.float32
masks.shape: (16, 128, 128, 1) images.dtype: tf.float32

Helper function for visualizing the results from TFRecords

def plot_masks(display_list):
  plt.figure(figsize=(15, 15))

  title = ['Input Image', 'True Mask', 'Predicted Mask']

  for i in range(len(display_list)):
    plt.subplot(1, len(display_list), i+1)
    plt.title(title[i])
    plt.imshow(tf.keras.utils.array_to_img(display_list[i]))


    plt.axis('off')
  plt.show()

Visualization of training data

Image Title represents what is depicted from the image.

Same helper function can be used while visualizing predicted mask

num_examples = 3

for images, masks in task.build_inputs(exp_config.task.train_data).take(num_examples):
  plot_masks([images[0], masks['masks'][0]])

png

png

png

Train and evaluate

IoU: is defined as the area of the intersection divided by the area of the union of a predicted mask and ground truth mask.

model, eval_logs = tfm.core.train_lib.run_experiment(
    distribution_strategy=distribution_strategy,
    task=task,
    mode='train_and_eval',
    params=exp_config,
    model_dir=model_dir,
    eval_summary_manager=summary_manager.maybe_build_eval_summary_manager(
              params=exp_config, model_dir=model_dir),
    run_post_eval=True)

INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
restoring or initializing model...
INFO:tensorflow:Customized initialization is done through the passed `init_fn`.
INFO:tensorflow:Customized initialization is done through the passed `init_fn`.
train | step:      0 | training until step 344...
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
WARNING:tensorflow:Gradients do not exist for variables ['conv2dbn_block_1/batch_normalization/gamma:0', 'conv2dbn_block_1/batch_normalization/beta:0'] when minimizing the loss. If you're using `model.compile()`, did you forget to provide a `loss` argument?
WARNING:tensorflow:Gradients do not exist for variables ['conv2dbn_block_1/batch_normalization/gamma:0', 'conv2dbn_block_1/batch_normalization/beta:0'] when minimizing the loss. If you're using `model.compile()`, did you forget to provide a `loss` argument?
WARNING:tensorflow:Gradients do not exist for variables ['conv2dbn_block_1/batch_normalization/gamma:0', 'conv2dbn_block_1/batch_normalization/beta:0'] when minimizing the loss. If you're using `model.compile()`, did you forget to provide a `loss` argument?
WARNING:tensorflow:Gradients do not exist for variables ['conv2dbn_block_1/batch_normalization/gamma:0', 'conv2dbn_block_1/batch_normalization/beta:0'] when minimizing the loss. If you're using `model.compile()`, did you forget to provide a `loss` argument?
WARNING:tensorflow:Gradients do not exist for variables ['conv2dbn_block_1/batch_normalization/gamma:0', 'conv2dbn_block_1/batch_normalization/beta:0'] when minimizing the loss. If you're using `model.compile()`, did you forget to provide a `loss` argument?
WARNING:tensorflow:Gradients do not exist for variables ['conv2dbn_block_1/batch_normalization/gamma:0', 'conv2dbn_block_1/batch_normalization/beta:0'] when minimizing the loss. If you're using `model.compile()`, did you forget to provide a `loss` argument?
WARNING:tensorflow:Gradients do not exist for variables ['conv2dbn_block_1/batch_normalization/gamma:0', 'conv2dbn_block_1/batch_normalization/beta:0'] when minimizing the loss. If you're using `model.compile()`, did you forget to provide a `loss` argument?
WARNING:tensorflow:Gradients do not exist for variables ['conv2dbn_block_1/batch_normalization/gamma:0', 'conv2dbn_block_1/batch_normalization/beta:0'] when minimizing the loss. If you're using `model.compile()`, did you forget to provide a `loss` argument?
INFO:tensorflow:Collective all_reduce tensors: 165 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 165 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
WARNING:tensorflow:Gradients do not exist for variables ['conv2dbn_block_1/batch_normalization/gamma:0', 'conv2dbn_block_1/batch_normalization/beta:0'] when minimizing the loss. If you're using `model.compile()`, did you forget to provide a `loss` argument?
WARNING:tensorflow:Gradients do not exist for variables ['conv2dbn_block_1/batch_normalization/gamma:0', 'conv2dbn_block_1/batch_normalization/beta:0'] when minimizing the loss. If you're using `model.compile()`, did you forget to provide a `loss` argument?
WARNING:tensorflow:Gradients do not exist for variables ['conv2dbn_block_1/batch_normalization/gamma:0', 'conv2dbn_block_1/batch_normalization/beta:0'] when minimizing the loss. If you're using `model.compile()`, did you forget to provide a `loss` argument?
WARNING:tensorflow:Gradients do not exist for variables ['conv2dbn_block_1/batch_normalization/gamma:0', 'conv2dbn_block_1/batch_normalization/beta:0'] when minimizing the loss. If you're using `model.compile()`, did you forget to provide a `loss` argument?
WARNING:tensorflow:Gradients do not exist for variables ['conv2dbn_block_1/batch_normalization/gamma:0', 'conv2dbn_block_1/batch_normalization/beta:0'] when minimizing the loss. If you're using `model.compile()`, did you forget to provide a `loss` argument?
WARNING:tensorflow:Gradients do not exist for variables ['conv2dbn_block_1/batch_normalization/gamma:0', 'conv2dbn_block_1/batch_normalization/beta:0'] when minimizing the loss. If you're using `model.compile()`, did you forget to provide a `loss` argument?
WARNING:tensorflow:Gradients do not exist for variables ['conv2dbn_block_1/batch_normalization/gamma:0', 'conv2dbn_block_1/batch_normalization/beta:0'] when minimizing the loss. If you're using `model.compile()`, did you forget to provide a `loss` argument?
WARNING:tensorflow:Gradients do not exist for variables ['conv2dbn_block_1/batch_normalization/gamma:0', 'conv2dbn_block_1/batch_normalization/beta:0'] when minimizing the loss. If you're using `model.compile()`, did you forget to provide a `loss` argument?
INFO:tensorflow:Collective all_reduce tensors: 165 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 165 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
train | step:    344 | steps/sec:    2.0 | output: 
    {'learning_rate': 0.09287634,
     'mean_iou': 0.5759032,
     'training_loss': 0.50292987}
saved checkpoint to ./trained_model/ckpt-344.
 eval | step:    344 | running 344 steps of evaluation...
WARNING:tensorflow:Setting parallel_iterations > 1 has no effect when executing eagerly. Consider calling map_fn with tf.function to execute fn in parallel.
WARNING:tensorflow:Setting parallel_iterations > 1 has no effect when executing eagerly. Consider calling map_fn with tf.function to execute fn in parallel.
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.9/site-packages/official/vision/utils/object_detection/visualization_utils.py:1016: calling map_fn_v2 (from tensorflow.python.ops.map_fn) with back_prop=False is deprecated and will be removed in a future version.
Instructions for updating:
back_prop=False is deprecated. Consider using tf.stop_gradient instead.
Instead of:
results = tf.map_fn(fn, elems, back_prop=False)
Use:
results = tf.nest.map_structure(tf.stop_gradient, tf.map_fn(fn, elems))
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.9/site-packages/official/vision/utils/object_detection/visualization_utils.py:1016: calling map_fn_v2 (from tensorflow.python.ops.map_fn) with back_prop=False is deprecated and will be removed in a future version.
Instructions for updating:
back_prop=False is deprecated. Consider using tf.stop_gradient instead.
Instead of:
results = tf.map_fn(fn, elems, back_prop=False)
Use:
results = tf.nest.map_structure(tf.stop_gradient, tf.map_fn(fn, elems))
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.9/site-packages/tensorflow/python/autograph/impl/api.py:460: py_func (from tensorflow.python.ops.script_ops) is deprecated and will be removed in a future version.
Instructions for updating:
tf.py_func is deprecated in TF V2. Instead, there are two
    options available in V2.

    - tf.py_function takes a python function which manipulates tf eager
    tensors instead of numpy arrays. It's easy to convert a tf eager tensor to
    an ndarray (just call tensor.numpy()) but having access to eager tensors
    means `tf.py_function`s can use accelerators such as GPUs as well as
    being differentiable using a gradient tape.
    - tf.numpy_function maintains the semantics of the deprecated tf.py_func
    (it is not differentiable, and manipulates numpy arrays). It drops the
    stateful argument making all functions stateful.
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.9/site-packages/tensorflow/python/autograph/impl/api.py:460: py_func (from tensorflow.python.ops.script_ops) is deprecated and will be removed in a future version.
Instructions for updating:
tf.py_func is deprecated in TF V2. Instead, there are two
    options available in V2.

    - tf.py_function takes a python function which manipulates tf eager
    tensors instead of numpy arrays. It's easy to convert a tf eager tensor to
    an ndarray (just call tensor.numpy()) but having access to eager tensors
    means `tf.py_function`s can use accelerators such as GPUs as well as
    being differentiable using a gradient tape.
    - tf.numpy_function maintains the semantics of the deprecated tf.py_func
    (it is not differentiable, and manipulates numpy arrays). It drops the
    stateful argument making all functions stateful.
eval | step:    344 | steps/sec:   30.2 | eval time:   11.4 sec | output: 
    {'best_mean_iou': 0.6224187,
     'image/validation_outputs/0': array([[[[120, 173,  62],
             [129, 175,  65],
             [129, 175,  67],
             ...,
             [136, 185,  80],
             [137, 186,  81],
             [137, 185,  78]],
    
            [[122, 175,  60],
             [128, 179,  65],
             [127, 179,  64],
             ...,
             [133, 185,  79],
             [138, 189,  85],
             [143, 189,  85]],
    
            [[105, 152,  30],
             [108, 155,  33],
             [105, 149,  26],
             ...,
             [144, 193,  88],
             [143, 196,  89],
             [141, 194,  88]],
    
            ...,
    
            [[  0, 102, 102],
             [  0, 102, 102],
             [  0, 102, 102],
             ...,
             [104, 164,  58],
             [101, 163,  56],
             [101, 160,  53]],
    
            [[  0, 102, 102],
             [  0, 102, 102],
             [  0, 102, 102],
             ...,
             [104, 165,  57],
             [101, 163,  55],
             [ 93, 152,  44]],
    
            [[  0, 102, 102],
             [  0, 102, 102],
             [  0, 102, 102],
             ...,
             [100, 164,  55],
             [ 89, 149,  41],
             [ 59, 110,   5]]]], dtype=uint8),
     'image/validation_outputs/1': array([[[[129, 190, 105],
             [130, 190, 106],
             [131, 191, 106],
             ...,
             [169, 228, 137],
             [168, 227, 137],
             [165, 226, 135]],
    
            [[131, 191, 107],
             [132, 192, 107],
             [132, 193, 108],
             ...,
             [170, 229, 138],
             [169, 228, 137],
             [168, 228, 137]],
    
            [[131, 191, 107],
             [132, 192, 107],
             [132, 193, 108],
             ...,
             [170, 229, 139],
             [170, 229, 139],
             [167, 227, 136]],
    
            ...,
    
            [[ 81, 124,  23],
             [ 81, 123,  23],
             [ 81, 124,  23],
             ...,
             [ 77, 119,  20],
             [ 77, 118,  20],
             [ 77, 119,  20]],
    
            [[ 80, 122,  22],
             [ 80, 122,  22],
             [ 83, 125,  25],
             ...,
             [ 77, 118,  20],
             [ 77, 118,  20],
             [ 75, 118,  19]],
    
            [[ 77, 121,  19],
             [ 78, 121,  19],
             [ 81, 124,  22],
             ...,
             [ 77, 119,  20],
             [ 77, 118,  20],
             [ 77, 119,  20]]]], dtype=uint8),
     'image/validation_outputs/10': array([[[[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            ...,
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]]]], dtype=uint8),
     'image/validation_outputs/11': array([[[[126, 176,  29],
             [130, 177,  34],
             [128, 176,  34],
             ...,
             [123, 174,  27],
             [121, 172,  26],
             [120, 171,  26]],
    
            [[127, 176,  32],
             [126, 172,  31],
             [111, 157,  17],
             ...,
             [120, 173,  26],
             [120, 172,  26],
             [120, 172,  27]],
    
            [[124, 175,  28],
             [126, 173,  30],
             [126, 175,  33],
             ...,
             [119, 173,  25],
             [117, 172,  24],
             [119, 172,  25]],
    
            ...,
    
            [[ 99, 161,  17],
             [104, 163,  23],
             [107, 163,  19],
             ...,
             [107, 166,  22],
             [ 98, 158,  13],
             [100, 163,  14]],
    
            [[ 98, 164,  15],
             [ 99, 164,  19],
             [ 98, 159,  11],
             ...,
             [113, 166,  28],
             [110, 163,  25],
             [105, 164,  20]],
    
            [[105, 167,  20],
             [ 97, 160,  14],
             [ 98, 154,  10],
             ...,
             [109, 164,  24],
             [110, 163,  28],
             [109, 162,  22]]]], dtype=uint8),
     'image/validation_outputs/12': array([[[[125, 166,  55],
             [123, 166,  55],
             [126, 169,  60],
             ...,
             [130, 167,  54],
             [129, 166,  55],
             [128, 163,  52]],
    
            [[123, 167,  55],
             [123, 167,  56],
             [122, 166,  56],
             ...,
             [127, 165,  52],
             [126, 163,  52],
             [126, 161,  50]],
    
            [[122, 166,  54],
             [122, 166,  55],
             [101, 145,  36],
             ...,
             [125, 163,  50],
             [125, 161,  51],
             [126, 161,  50]],
    
            ...,
    
            [[116, 159,  49],
             [109, 151,  41],
             [116, 160,  51],
             ...,
             [107, 150,  41],
             [111, 155,  46],
             [120, 164,  55]],
    
            [[108, 151,  41],
             [101, 144,  35],
             [112, 156,  47],
             ...,
             [115, 160,  51],
             [117, 161,  53],
             [126, 171,  61]],
    
            [[104, 147,  38],
             [103, 145,  37],
             [107, 151,  42],
             ...,
             [121, 166,  55],
             [117, 161,  52],
             [130, 175,  64]]]], dtype=uint8),
     'image/validation_outputs/2': array([[[[188, 250, 149],
             [192, 249, 149],
             [197, 252, 151],
             ...,
             [ 51, 114,  19],
             [ 52, 115,  21],
             [ 51, 114,  19]],
    
            [[191, 251, 151],
             [186, 250, 151],
             [185, 245, 144],
             ...,
             [ 52, 115,  20],
             [ 52, 115,  22],
             [ 51, 115,  20]],
    
            [[185, 245, 147],
             [177, 245, 146],
             [138, 203, 104],
             ...,
             [ 52, 115,  20],
             [ 52, 115,  22],
             [ 52, 115,  22]],
    
            ...,
    
            [[134, 206, 114],
             [135, 205, 113],
             [135, 206, 114],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[134, 205, 113],
             [135, 204, 113],
             [134, 203, 112],
             ...,
             [ 51, 102,   0],
             [ 52, 102,   0],
             [ 52, 102,   0]],
    
            [[132, 202, 110],
             [133, 202, 110],
             [132, 202, 109],
             ...,
             [ 51, 102,   0],
             [ 52, 102,   0],
             [ 52, 102,   0]]]], dtype=uint8),
     'image/validation_outputs/3': array([[[[ 59, 104,   1],
             [ 51, 103,   0],
             [ 66, 112,   7],
             ...,
             [ 68, 120,   8],
             [ 71, 122,  12],
             [ 73, 123,  12]],
    
            [[ 58, 104,   0],
             [ 51, 103,   0],
             [ 65, 112,   6],
             ...,
             [ 68, 120,  10],
             [ 71, 123,  13],
             [ 72, 124,  12]],
    
            [[ 59, 105,   1],
             [ 51, 103,   0],
             [ 60, 107,   1],
             ...,
             [ 66, 118,   8],
             [ 68, 119,  10],
             [ 68, 119,   9]],
    
            ...,
    
            [[ 55, 106,   0],
             [ 55, 106,   0],
             [ 46, 103,   5],
             ...,
             [ 62, 112,   1],
             [ 63, 112,   2],
             [ 63, 110,   2]],
    
            [[ 55, 106,   0],
             [ 55, 107,   0],
             [ 47, 104,   6],
             ...,
             [ 65, 112,   2],
             [ 63, 110,   1],
             [ 65, 111,   1]],
    
            [[ 54, 105,   0],
             [ 54, 106,   0],
             [ 48, 106,   7],
             ...,
             [ 63, 110,   2],
             [ 63, 109,   2],
             [ 63, 110,   0]]]], dtype=uint8),
     'image/validation_outputs/4': array([[[[145, 188,  78],
             [145, 189,  80],
             [144, 188,  79],
             ...,
             [140, 179,  67],
             [143, 182,  72],
             [141, 185,  70]],
    
            [[143, 187,  79],
             [143, 187,  79],
             [140, 184,  76],
             ...,
             [134, 176,  65],
             [140, 184,  71],
             [139, 182,  70]],
    
            [[143, 187,  79],
             [147, 190,  82],
             [120, 163,  56],
             ...,
             [125, 167,  58],
             [125, 169,  57],
             [123, 170,  58]],
    
            ...,
    
            [[137, 175,  64],
             [129, 170,  57],
             [131, 170,  61],
             ...,
             [136, 178,  68],
             [132, 175,  63],
             [118, 160,  49]],
    
            [[ 95, 136,  21],
             [140, 182,  70],
             [140, 181,  73],
             ...,
             [129, 170,  61],
             [128, 170,  59],
             [130, 172,  60]],
    
            [[ 95, 139,  21],
             [138, 184,  72],
             [142, 184,  77],
             ...,
             [150, 190,  82],
             [151, 194,  83],
             [140, 181,  71]]]], dtype=uint8),
     'image/validation_outputs/5': array([[[[ 74, 119,  12],
             [ 71, 118,  10],
             [ 65, 113,   7],
             ...,
             [ 63, 118,   4],
             [ 68, 122,   5],
             [ 60, 115,   2]],
    
            [[ 77, 121,  14],
             [ 78, 121,  13],
             [ 77, 120,  12],
             ...,
             [ 63, 118,   4],
             [ 77, 126,   9],
             [ 54, 107,   0]],
    
            [[ 65, 113,   6],
             [ 83, 127,  19],
             [ 78, 121,  12],
             ...,
             [ 60, 118,   5],
             [ 72, 122,   9],
             [ 62, 112,   5]],
    
            ...,
    
            [[141, 206, 112],
             [141, 206, 113],
             [142, 208, 112],
             ...,
             [149, 211, 115],
             [147, 209, 113],
             [144, 208, 111]],
    
            [[134, 203, 110],
             [135, 203, 112],
             [136, 205, 112],
             ...,
             [140, 204, 110],
             [138, 203, 108],
             [135, 203, 108]],
    
            [[123, 196, 105],
             [124, 196, 107],
             [125, 198, 107],
             ...,
             [140, 202, 109],
             [138, 203, 109],
             [138, 202, 107]]]], dtype=uint8),
     'image/validation_outputs/6': array([[[[147, 212,  98],
             [174, 236, 126],
             [169, 229, 121],
             ...,
             [ 57, 121,  17],
             [132, 195,  94],
             [ 95, 155,  58]],
    
            [[141, 209,  96],
             [153, 217, 109],
             [132, 192,  83],
             ...,
             [ 96, 159,  59],
             [ 86, 149,  50],
             [137, 199,  95]],
    
            [[152, 212, 109],
             [180, 239, 137],
             [158, 218, 111],
             ...,
             [187, 243, 147],
             [189, 245, 146],
             [142, 203,  98]],
    
            ...,
    
            [[ 96, 181,  22],
             [ 93, 181,  11],
             [ 80, 175,   2],
             ...,
             [174, 225, 113],
             [173, 224, 113],
             [167, 218, 105]],
    
            [[102, 189,  27],
             [100, 183,  19],
             [ 85, 170,   5],
             ...,
             [164, 214, 103],
             [172, 222, 112],
             [168, 220, 108]],
    
            [[104, 181,  26],
             [114, 177,  32],
             [ 75, 159,   0],
             ...,
             [170, 220, 110],
             [168, 220, 110],
             [165, 217, 106]]]], dtype=uint8),
     'image/validation_outputs/7': array([[[[161, 172,  20],
             [156, 172,  18],
             [157, 170,  21],
             ...,
             [182, 218,  90],
             [181, 217,  90],
             [179, 217,  87]],
    
            [[156, 173,  17],
             [154, 170,  21],
             [145, 166,  23],
             ...,
             [182, 219,  91],
             [182, 218,  90],
             [182, 219,  89]],
    
            [[156, 170,  16],
             [158, 170,  21],
             [149, 169,  25],
             ...,
             [182, 220,  91],
             [182, 219,  91],
             [180, 219,  90]],
    
            ...,
    
            [[158, 169,  22],
             [161, 173,  25],
             [161, 175,  26],
             ...,
             [174, 199,  56],
             [171, 194,  53],
             [170, 194,  50]],
    
            [[158, 169,  21],
             [160, 172,  21],
             [161, 173,  23],
             ...,
             [172, 198,  58],
             [168, 196,  56],
             [170, 197,  55]],
    
            [[156, 167,  21],
             [158, 170,  22],
             [161, 173,  24],
             ...,
             [170, 196,  56],
             [169, 197,  56],
             [172, 201,  62]]]], dtype=uint8),
     'image/validation_outputs/8': array([[[[117, 151,  19],
             [119, 152,  19],
             [119, 152,  20],
             ...,
             [116, 150,  22],
             [118, 150,  23],
             [114, 149,  20]],
    
            [[117, 151,  19],
             [119, 152,  22],
             [114, 148,  20],
             ...,
             [119, 152,  25],
             [116, 148,  22],
             [113, 148,  23]],
    
            [[120, 152,  21],
             [120, 153,  23],
             [ 98, 131,   5],
             ...,
             [122, 154,  25],
             [117, 149,  23],
             [114, 148,  22]],
    
            ...,
    
            [[122, 155,  23],
             [111, 144,  17],
             [111, 146,  17],
             ...,
             [126, 158,  33],
             [132, 163,  33],
             [126, 158,  31]],
    
            [[116, 148,  17],
             [108, 142,  16],
             [113, 147,  19],
             ...,
             [127, 158,  33],
             [127, 158,  30],
             [128, 159,  31]],
    
            [[113, 147,  16],
             [108, 143,  14],
             [109, 143,  13],
             ...,
             [122, 155,  30],
             [125, 156,  29],
             [125, 156,  30]]]], dtype=uint8),
     'image/validation_outputs/9': array([[[[160, 209,  91],
             [159, 209,  92],
             [160, 212,  94],
             ...,
             [146, 191,  52],
             [144, 189,  50],
             [143, 188,  49]],
    
            [[159, 209,  90],
             [159, 209,  92],
             [161, 212,  95],
             ...,
             [144, 190,  50],
             [143, 188,  49],
             [144, 189,  50]],
    
            [[160, 210,  91],
             [158, 209,  91],
             [160, 212,  94],
             ...,
             [146, 191,  52],
             [144, 190,  50],
             [143, 188,  49]],
    
            ...,
    
            [[124, 174,  37],
             [124, 174,  37],
             [125, 174,  37],
             ...,
             [111, 159,  20],
             [109, 158,  19],
             [109, 157,  19]],
    
            [[124, 174,  35],
             [124, 174,  37],
             [126, 175,  38],
             ...,
             [112, 161,  22],
             [111, 159,  20],
             [109, 158,  19]],
    
            [[125, 175,  36],
             [126, 176,  39],
             [128, 178,  41],
             ...,
             [111, 159,  20],
             [110, 158,  19],
             [109, 157,  18]]]], dtype=uint8),
     'iou/0': 0.7160484,
     'iou/1': 0.8467605,
     'iou/2': 0.30444714,
     'mean_iou': 0.6224187,
     'steps_per_second': 30.152094229540886,
     'validation_loss': 0.49941143}
train | step:    344 | training until step 688...
train | step:    688 | steps/sec:    5.2 | output: 
    {'learning_rate': 0.0735352,
     'mean_iou': 0.63586026,
     'training_loss': 0.42186454}
saved checkpoint to ./trained_model/ckpt-688.
 eval | step:    688 | running 344 steps of evaluation...
 eval | step:    688 | steps/sec:  108.5 | eval time:    3.2 sec | output: 
    {'best_mean_iou': 0.6573053,
     'image/validation_outputs/0': array([[[[120, 173,  62],
             [129, 175,  65],
             [129, 175,  67],
             ...,
             [136, 185,  80],
             [137, 186,  81],
             [137, 185,  78]],
    
            [[122, 175,  60],
             [128, 179,  65],
             [127, 179,  64],
             ...,
             [133, 185,  79],
             [138, 189,  85],
             [143, 189,  85]],
    
            [[105, 152,  30],
             [108, 155,  33],
             [105, 149,  26],
             ...,
             [144, 193,  88],
             [143, 196,  89],
             [141, 194,  88]],
    
            ...,
    
            [[ 45, 102,   0],
             [ 45, 102,   0],
             [ 46, 102,   0],
             ...,
             [104, 164,  58],
             [101, 163,  56],
             [101, 160,  53]],
    
            [[ 44, 102,   0],
             [ 45, 102,   0],
             [ 46, 102,   0],
             ...,
             [104, 165,  57],
             [101, 163,  55],
             [ 93, 152,  44]],
    
            [[ 44, 102,   0],
             [ 45, 102,   0],
             [ 46, 102,   0],
             ...,
             [100, 164,  55],
             [ 89, 149,  41],
             [ 59, 110,   5]]]], dtype=uint8),
     'image/validation_outputs/1': array([[[[129, 190, 105],
             [130, 190, 106],
             [131, 191, 106],
             ...,
             [169, 228, 137],
             [168, 227, 137],
             [165, 226, 135]],
    
            [[131, 191, 107],
             [132, 192, 107],
             [132, 193, 108],
             ...,
             [170, 229, 138],
             [169, 228, 137],
             [168, 228, 137]],
    
            [[131, 191, 107],
             [132, 192, 107],
             [132, 193, 108],
             ...,
             [170, 229, 139],
             [170, 229, 139],
             [167, 227, 136]],
    
            ...,
    
            [[ 81, 124,  23],
             [ 81, 123,  23],
             [ 81, 124,  23],
             ...,
             [ 77, 119,  20],
             [ 77, 118,  20],
             [ 77, 119,  20]],
    
            [[ 80, 122,  22],
             [ 80, 122,  22],
             [ 83, 125,  25],
             ...,
             [ 77, 118,  20],
             [ 77, 118,  20],
             [ 75, 118,  19]],
    
            [[ 77, 121,  19],
             [ 78, 121,  19],
             [ 81, 124,  22],
             ...,
             [ 77, 119,  20],
             [ 77, 118,  20],
             [ 77, 119,  20]]]], dtype=uint8),
     'image/validation_outputs/10': array([[[[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            ...,
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]]]], dtype=uint8),
     'image/validation_outputs/11': array([[[[126, 176,  29],
             [130, 177,  34],
             [128, 176,  34],
             ...,
             [122, 174,  27],
             [120, 172,  26],
             [119, 171,  26]],
    
            [[127, 176,  32],
             [126, 172,  31],
             [111, 157,  17],
             ...,
             [119, 173,  26],
             [120, 172,  26],
             [120, 172,  27]],
    
            [[124, 175,  28],
             [126, 173,  30],
             [126, 175,  33],
             ...,
             [118, 173,  25],
             [117, 172,  24],
             [118, 172,  25]],
    
            ...,
    
            [[100, 161,  17],
             [106, 163,  23],
             [108, 163,  19],
             ...,
             [107, 166,  22],
             [ 98, 158,  13],
             [100, 163,  14]],
    
            [[ 99, 164,  15],
             [100, 164,  19],
             [ 99, 159,  11],
             ...,
             [113, 166,  28],
             [110, 163,  25],
             [105, 164,  20]],
    
            [[106, 167,  20],
             [ 98, 160,  14],
             [ 99, 154,  10],
             ...,
             [109, 164,  24],
             [110, 163,  28],
             [109, 162,  22]]]], dtype=uint8),
     'image/validation_outputs/12': array([[[[125, 166,  55],
             [123, 166,  55],
             [126, 169,  60],
             ...,
             [130, 167,  54],
             [129, 166,  55],
             [128, 163,  52]],
    
            [[123, 167,  55],
             [123, 167,  56],
             [122, 166,  56],
             ...,
             [127, 165,  52],
             [126, 163,  52],
             [126, 161,  50]],
    
            [[122, 166,  54],
             [122, 166,  55],
             [101, 145,  36],
             ...,
             [125, 163,  50],
             [125, 161,  51],
             [126, 161,  50]],
    
            ...,
    
            [[116, 159,  49],
             [109, 151,  41],
             [116, 160,  51],
             ...,
             [107, 150,  41],
             [111, 155,  46],
             [120, 164,  55]],
    
            [[108, 151,  41],
             [101, 144,  35],
             [112, 156,  47],
             ...,
             [115, 160,  51],
             [117, 161,  53],
             [126, 171,  61]],
    
            [[104, 147,  38],
             [103, 145,  37],
             [107, 151,  42],
             ...,
             [121, 166,  55],
             [117, 161,  52],
             [130, 175,  64]]]], dtype=uint8),
     'image/validation_outputs/2': array([[[[188, 250, 149],
             [192, 249, 149],
             [197, 252, 151],
             ...,
             [ 52, 114,  19],
             [ 53, 115,  21],
             [ 53, 114,  19]],
    
            [[191, 251, 151],
             [186, 250, 151],
             [185, 245, 144],
             ...,
             [ 54, 115,  20],
             [ 54, 115,  22],
             [ 53, 115,  20]],
    
            [[185, 245, 147],
             [177, 245, 146],
             [138, 203, 104],
             ...,
             [ 52, 115,  20],
             [ 54, 115,  22],
             [ 53, 115,  22]],
    
            ...,
    
            [[134, 206, 114],
             [135, 205, 113],
             [135, 206, 114],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[134, 205, 113],
             [135, 204, 113],
             [134, 203, 112],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[132, 202, 110],
             [133, 202, 110],
             [132, 202, 109],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]]]], dtype=uint8),
     'image/validation_outputs/3': array([[[[ 59, 104,   1],
             [ 51, 103,   0],
             [ 66, 112,   7],
             ...,
             [ 68, 120,   8],
             [ 71, 122,  12],
             [ 73, 123,  12]],
    
            [[ 58, 104,   0],
             [ 51, 103,   0],
             [ 65, 112,   6],
             ...,
             [ 68, 120,  10],
             [ 71, 123,  13],
             [ 72, 124,  12]],
    
            [[ 59, 105,   1],
             [ 51, 103,   0],
             [ 60, 107,   1],
             ...,
             [ 66, 118,   8],
             [ 68, 119,  10],
             [ 68, 119,   9]],
    
            ...,
    
            [[ 55, 106,   0],
             [ 54, 106,   0],
             [ 46, 103,   5],
             ...,
             [ 62, 112,   1],
             [ 62, 112,   2],
             [ 62, 110,   2]],
    
            [[ 54, 106,   0],
             [ 54, 107,   0],
             [ 46, 104,   6],
             ...,
             [ 64, 112,   2],
             [ 62, 110,   1],
             [ 64, 111,   1]],
    
            [[ 53, 105,   0],
             [ 53, 106,   0],
             [ 47, 106,   7],
             ...,
             [ 62, 110,   2],
             [ 62, 109,   2],
             [ 62, 110,   0]]]], dtype=uint8),
     'image/validation_outputs/4': array([[[[145, 188,  78],
             [145, 189,  80],
             [144, 188,  79],
             ...,
             [140, 179,  67],
             [143, 182,  72],
             [141, 185,  70]],
    
            [[143, 187,  79],
             [143, 187,  79],
             [140, 184,  76],
             ...,
             [134, 176,  65],
             [140, 184,  71],
             [139, 182,  70]],
    
            [[143, 187,  79],
             [147, 190,  82],
             [120, 163,  56],
             ...,
             [125, 167,  58],
             [125, 169,  57],
             [123, 170,  58]],
    
            ...,
    
            [[137, 175,  64],
             [129, 170,  57],
             [131, 170,  61],
             ...,
             [136, 178,  68],
             [132, 175,  63],
             [118, 160,  49]],
    
            [[ 95, 136,  21],
             [140, 182,  70],
             [140, 181,  73],
             ...,
             [129, 170,  61],
             [128, 170,  59],
             [130, 172,  60]],
    
            [[ 95, 139,  21],
             [138, 184,  72],
             [142, 184,  77],
             ...,
             [150, 190,  82],
             [151, 194,  83],
             [140, 181,  71]]]], dtype=uint8),
     'image/validation_outputs/5': array([[[[ 74, 119,  12],
             [ 71, 118,  10],
             [ 65, 113,   7],
             ...,
             [ 63, 118,   4],
             [ 68, 122,   5],
             [ 60, 115,   2]],
    
            [[ 77, 121,  14],
             [ 78, 121,  13],
             [ 77, 120,  12],
             ...,
             [ 63, 118,   4],
             [ 77, 126,   9],
             [ 54, 107,   0]],
    
            [[ 65, 113,   6],
             [ 83, 127,  19],
             [ 78, 121,  12],
             ...,
             [ 60, 118,   5],
             [ 72, 122,   9],
             [ 62, 112,   5]],
    
            ...,
    
            [[141, 206, 112],
             [141, 206, 113],
             [142, 208, 112],
             ...,
             [149, 211, 115],
             [147, 209, 113],
             [144, 208, 111]],
    
            [[134, 203, 110],
             [135, 203, 112],
             [136, 205, 112],
             ...,
             [140, 204, 110],
             [138, 203, 108],
             [135, 203, 108]],
    
            [[123, 196, 105],
             [124, 196, 107],
             [125, 198, 107],
             ...,
             [140, 202, 109],
             [138, 203, 109],
             [138, 202, 107]]]], dtype=uint8),
     'image/validation_outputs/6': array([[[[147, 212,  98],
             [174, 236, 126],
             [169, 229, 121],
             ...,
             [ 57, 121,  17],
             [132, 195,  94],
             [ 95, 155,  58]],
    
            [[141, 209,  96],
             [153, 217, 109],
             [132, 192,  83],
             ...,
             [ 96, 159,  59],
             [ 86, 149,  50],
             [137, 199,  95]],
    
            [[152, 212, 109],
             [180, 239, 137],
             [158, 218, 111],
             ...,
             [187, 243, 147],
             [189, 245, 146],
             [142, 203,  98]],
    
            ...,
    
            [[ 96, 181,  22],
             [ 93, 181,  11],
             [ 80, 175,   2],
             ...,
             [174, 225, 113],
             [173, 224, 113],
             [167, 218, 105]],
    
            [[102, 189,  27],
             [100, 183,  19],
             [ 85, 170,   5],
             ...,
             [164, 214, 103],
             [172, 222, 112],
             [168, 220, 108]],
    
            [[104, 181,  26],
             [114, 177,  32],
             [ 75, 159,   0],
             ...,
             [170, 220, 110],
             [168, 220, 110],
             [165, 217, 106]]]], dtype=uint8),
     'image/validation_outputs/7': array([[[[161, 172,  20],
             [156, 172,  18],
             [157, 170,  21],
             ...,
             [182, 218,  90],
             [181, 217,  90],
             [179, 217,  87]],
    
            [[156, 173,  17],
             [154, 170,  21],
             [145, 166,  23],
             ...,
             [182, 219,  91],
             [182, 218,  90],
             [182, 219,  89]],
    
            [[156, 170,  16],
             [158, 170,  21],
             [149, 169,  25],
             ...,
             [182, 220,  91],
             [182, 219,  91],
             [180, 219,  90]],
    
            ...,
    
            [[157, 169,  22],
             [160, 173,  25],
             [161, 175,  26],
             ...,
             [179, 199,  56],
             [176, 194,  53],
             [176, 194,  50]],
    
            [[157, 169,  21],
             [159, 172,  21],
             [160, 173,  23],
             ...,
             [177, 198,  58],
             [174, 196,  56],
             [176, 197,  55]],
    
            [[156, 167,  21],
             [157, 170,  22],
             [160, 173,  24],
             ...,
             [176, 196,  56],
             [175, 197,  56],
             [179, 201,  62]]]], dtype=uint8),
     'image/validation_outputs/8': array([[[[117, 151,  19],
             [119, 152,  19],
             [119, 152,  20],
             ...,
             [116, 150,  22],
             [118, 150,  23],
             [114, 149,  20]],
    
            [[117, 151,  19],
             [119, 152,  22],
             [114, 148,  20],
             ...,
             [119, 152,  25],
             [116, 148,  22],
             [113, 148,  23]],
    
            [[120, 152,  21],
             [120, 153,  23],
             [ 98, 131,   5],
             ...,
             [122, 154,  25],
             [117, 149,  23],
             [114, 148,  22]],
    
            ...,
    
            [[122, 155,  23],
             [111, 144,  17],
             [111, 146,  17],
             ...,
             [126, 158,  33],
             [132, 163,  33],
             [126, 158,  31]],
    
            [[116, 148,  17],
             [108, 142,  16],
             [113, 147,  19],
             ...,
             [127, 158,  33],
             [127, 158,  30],
             [128, 159,  31]],
    
            [[113, 147,  16],
             [108, 143,  14],
             [109, 143,  13],
             ...,
             [122, 155,  30],
             [125, 156,  29],
             [125, 156,  30]]]], dtype=uint8),
     'image/validation_outputs/9': array([[[[160, 209,  91],
             [159, 209,  92],
             [160, 212,  94],
             ...,
             [146, 191,  52],
             [144, 189,  50],
             [143, 188,  49]],
    
            [[159, 209,  90],
             [159, 209,  92],
             [161, 212,  95],
             ...,
             [144, 190,  50],
             [143, 188,  49],
             [144, 189,  50]],
    
            [[160, 210,  91],
             [158, 209,  91],
             [160, 212,  94],
             ...,
             [146, 191,  52],
             [144, 190,  50],
             [143, 188,  49]],
    
            ...,
    
            [[124, 174,  37],
             [124, 174,  37],
             [125, 174,  37],
             ...,
             [117, 159,  20],
             [116, 158,  19],
             [115, 157,  19]],
    
            [[124, 174,  35],
             [124, 174,  37],
             [126, 175,  38],
             ...,
             [119, 161,  22],
             [117, 159,  20],
             [116, 158,  19]],
    
            [[125, 175,  36],
             [126, 176,  39],
             [128, 178,  41],
             ...,
             [118, 159,  20],
             [117, 158,  19],
             [116, 157,  18]]]], dtype=uint8),
     'iou/0': 0.7739854,
     'iou/1': 0.8697517,
     'iou/2': 0.3281788,
     'mean_iou': 0.6573053,
     'steps_per_second': 108.47001408687251,
     'validation_loss': 0.40915582}
train | step:    688 | training until step 1032...
train | step:   1032 | steps/sec:    6.0 | output: 
    {'learning_rate': 0.047487784,
     'mean_iou': 0.67253137,
     'training_loss': 0.374695}
saved checkpoint to ./trained_model/ckpt-1032.
 eval | step:   1032 | running 344 steps of evaluation...
 eval | step:   1032 | steps/sec:   78.0 | eval time:    4.4 sec | output: 
    {'best_mean_iou': 0.67205024,
     'image/validation_outputs/0': array([[[[120, 173,  62],
             [129, 175,  65],
             [129, 175,  67],
             ...,
             [136, 185,  80],
             [137, 186,  81],
             [137, 185,  78]],
    
            [[122, 175,  60],
             [128, 179,  65],
             [127, 179,  64],
             ...,
             [133, 185,  79],
             [138, 189,  85],
             [143, 189,  85]],
    
            [[105, 152,  30],
             [108, 155,  33],
             [105, 149,  26],
             ...,
             [144, 193,  88],
             [143, 196,  89],
             [141, 194,  88]],
    
            ...,
    
            [[ 45, 102,   0],
             [ 46, 102,   0],
             [ 46, 102,   0],
             ...,
             [104, 164,  58],
             [101, 163,  56],
             [101, 160,  53]],
    
            [[ 45, 102,   0],
             [ 45, 102,   0],
             [ 45, 102,   0],
             ...,
             [104, 165,  57],
             [101, 163,  55],
             [ 93, 152,  44]],
    
            [[ 44, 102,   0],
             [ 45, 102,   0],
             [ 45, 102,   0],
             ...,
             [100, 164,  55],
             [ 89, 149,  41],
             [ 59, 110,   5]]]], dtype=uint8),
     'image/validation_outputs/1': array([[[[129, 190, 105],
             [130, 190, 106],
             [131, 191, 106],
             ...,
             [169, 228, 137],
             [168, 227, 137],
             [165, 226, 135]],
    
            [[131, 191, 107],
             [132, 192, 107],
             [132, 193, 108],
             ...,
             [170, 229, 138],
             [169, 228, 137],
             [168, 228, 137]],
    
            [[131, 191, 107],
             [132, 192, 107],
             [132, 193, 108],
             ...,
             [170, 229, 139],
             [170, 229, 139],
             [167, 227, 136]],
    
            ...,
    
            [[ 81, 124,  23],
             [ 81, 123,  23],
             [ 81, 124,  23],
             ...,
             [ 77, 119,  20],
             [ 77, 118,  20],
             [ 77, 119,  20]],
    
            [[ 80, 122,  22],
             [ 80, 122,  22],
             [ 83, 125,  25],
             ...,
             [ 77, 118,  20],
             [ 77, 118,  20],
             [ 75, 118,  19]],
    
            [[ 77, 121,  19],
             [ 78, 121,  19],
             [ 81, 124,  22],
             ...,
             [ 77, 119,  20],
             [ 77, 118,  20],
             [ 77, 119,  20]]]], dtype=uint8),
     'image/validation_outputs/10': array([[[[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            ...,
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]]]], dtype=uint8),
     'image/validation_outputs/11': array([[[[126, 176,  29],
             [130, 177,  34],
             [128, 176,  34],
             ...,
             [123, 174,  27],
             [121, 172,  26],
             [120, 171,  26]],
    
            [[127, 176,  32],
             [126, 172,  31],
             [111, 157,  17],
             ...,
             [120, 173,  26],
             [120, 172,  26],
             [120, 172,  27]],
    
            [[124, 175,  28],
             [126, 173,  30],
             [126, 175,  33],
             ...,
             [119, 173,  25],
             [117, 172,  24],
             [119, 172,  25]],
    
            ...,
    
            [[100, 161,  17],
             [106, 163,  23],
             [108, 163,  19],
             ...,
             [107, 166,  22],
             [ 98, 158,  13],
             [100, 163,  14]],
    
            [[ 99, 164,  15],
             [100, 164,  19],
             [ 99, 159,  11],
             ...,
             [113, 166,  28],
             [110, 163,  25],
             [105, 164,  20]],
    
            [[106, 167,  20],
             [ 98, 160,  14],
             [ 99, 154,  10],
             ...,
             [109, 164,  24],
             [110, 163,  28],
             [109, 162,  22]]]], dtype=uint8),
     'image/validation_outputs/12': array([[[[125, 166,  55],
             [123, 166,  55],
             [126, 169,  60],
             ...,
             [119, 167,  54],
             [118, 166,  55],
             [116, 163,  52]],
    
            [[123, 167,  55],
             [123, 167,  56],
             [122, 166,  56],
             ...,
             [117, 165,  52],
             [115, 163,  52],
             [114, 161,  50]],
    
            [[122, 166,  54],
             [122, 166,  55],
             [101, 145,  36],
             ...,
             [115, 163,  50],
             [113, 161,  51],
             [114, 161,  50]],
    
            ...,
    
            [[116, 159,  49],
             [109, 151,  41],
             [116, 160,  51],
             ...,
             [108, 150,  41],
             [112, 155,  46],
             [121, 164,  55]],
    
            [[108, 151,  41],
             [101, 144,  35],
             [112, 156,  47],
             ...,
             [117, 160,  51],
             [119, 161,  53],
             [128, 171,  61]],
    
            [[104, 147,  38],
             [103, 145,  37],
             [107, 151,  42],
             ...,
             [122, 166,  55],
             [119, 161,  52],
             [131, 175,  64]]]], dtype=uint8),
     'image/validation_outputs/2': array([[[[188, 250, 149],
             [192, 249, 149],
             [197, 252, 151],
             ...,
             [ 51, 114,  19],
             [ 53, 115,  21],
             [ 52, 114,  19]],
    
            [[191, 251, 151],
             [186, 250, 151],
             [185, 245, 144],
             ...,
             [ 53, 115,  20],
             [ 53, 115,  22],
             [ 52, 115,  20]],
    
            [[185, 245, 147],
             [177, 245, 146],
             [138, 203, 104],
             ...,
             [ 52, 115,  20],
             [ 53, 115,  22],
             [ 53, 115,  22]],
    
            ...,
    
            [[134, 206, 114],
             [135, 205, 113],
             [135, 206, 114],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[134, 205, 113],
             [135, 204, 113],
             [134, 203, 112],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[132, 202, 110],
             [133, 202, 110],
             [132, 202, 109],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]]]], dtype=uint8),
     'image/validation_outputs/3': array([[[[ 59, 104,   1],
             [ 51, 103,   0],
             [ 66, 112,   7],
             ...,
             [ 68, 120,   8],
             [ 71, 122,  12],
             [ 73, 123,  12]],
    
            [[ 58, 104,   0],
             [ 51, 103,   0],
             [ 65, 112,   6],
             ...,
             [ 68, 120,  10],
             [ 71, 123,  13],
             [ 72, 124,  12]],
    
            [[ 59, 105,   1],
             [ 51, 103,   0],
             [ 60, 107,   1],
             ...,
             [ 66, 118,   8],
             [ 68, 119,  10],
             [ 68, 119,   9]],
    
            ...,
    
            [[ 61, 106,   0],
             [ 60, 106,   0],
             [ 51, 103,   5],
             ...,
             [ 62, 112,   1],
             [ 62, 112,   2],
             [ 62, 110,   2]],
    
            [[ 61, 106,   0],
             [ 61, 107,   0],
             [ 52, 104,   6],
             ...,
             [ 64, 112,   2],
             [ 62, 110,   1],
             [ 64, 111,   1]],
    
            [[ 61, 105,   0],
             [ 60, 106,   0],
             [ 53, 106,   7],
             ...,
             [ 62, 110,   2],
             [ 62, 109,   2],
             [ 62, 110,   0]]]], dtype=uint8),
     'image/validation_outputs/4': array([[[[145, 188,  78],
             [145, 189,  80],
             [144, 188,  79],
             ...,
             [140, 179,  67],
             [143, 182,  72],
             [141, 185,  70]],
    
            [[143, 187,  79],
             [143, 187,  79],
             [140, 184,  76],
             ...,
             [134, 176,  65],
             [140, 184,  71],
             [139, 182,  70]],
    
            [[143, 187,  79],
             [147, 190,  82],
             [120, 163,  56],
             ...,
             [125, 167,  58],
             [125, 169,  57],
             [123, 170,  58]],
    
            ...,
    
            [[137, 175,  64],
             [129, 170,  57],
             [131, 170,  61],
             ...,
             [136, 178,  68],
             [132, 175,  63],
             [118, 160,  49]],
    
            [[ 95, 136,  21],
             [140, 182,  70],
             [140, 181,  73],
             ...,
             [129, 170,  61],
             [128, 170,  59],
             [130, 172,  60]],
    
            [[ 95, 139,  21],
             [138, 184,  72],
             [142, 184,  77],
             ...,
             [150, 190,  82],
             [151, 194,  83],
             [140, 181,  71]]]], dtype=uint8),
     'image/validation_outputs/5': array([[[[ 74, 119,  12],
             [ 71, 118,  10],
             [ 65, 113,   7],
             ...,
             [ 63, 118,   4],
             [ 68, 122,   5],
             [ 60, 115,   2]],
    
            [[ 77, 121,  14],
             [ 78, 121,  13],
             [ 77, 120,  12],
             ...,
             [ 63, 118,   4],
             [ 77, 126,   9],
             [ 54, 107,   0]],
    
            [[ 65, 113,   6],
             [ 83, 127,  19],
             [ 78, 121,  12],
             ...,
             [ 60, 118,   5],
             [ 72, 122,   9],
             [ 62, 112,   5]],
    
            ...,
    
            [[141, 206, 112],
             [141, 206, 113],
             [142, 208, 112],
             ...,
             [149, 211, 115],
             [147, 209, 113],
             [144, 208, 111]],
    
            [[134, 203, 110],
             [135, 203, 112],
             [136, 205, 112],
             ...,
             [140, 204, 110],
             [138, 203, 108],
             [135, 203, 108]],
    
            [[123, 196, 105],
             [124, 196, 107],
             [125, 198, 107],
             ...,
             [140, 202, 109],
             [138, 203, 109],
             [138, 202, 107]]]], dtype=uint8),
     'image/validation_outputs/6': array([[[[147, 212,  98],
             [174, 236, 126],
             [169, 229, 121],
             ...,
             [ 57, 121,  17],
             [132, 195,  94],
             [ 95, 155,  58]],
    
            [[141, 209,  96],
             [153, 217, 109],
             [132, 192,  83],
             ...,
             [ 96, 159,  59],
             [ 86, 149,  50],
             [137, 199,  95]],
    
            [[152, 212, 109],
             [180, 239, 137],
             [158, 218, 111],
             ...,
             [187, 243, 147],
             [189, 245, 146],
             [142, 203,  98]],
    
            ...,
    
            [[ 96, 181,  22],
             [ 93, 181,  11],
             [ 80, 175,   2],
             ...,
             [174, 225, 113],
             [173, 224, 113],
             [167, 218, 105]],
    
            [[102, 189,  27],
             [100, 183,  19],
             [ 85, 170,   5],
             ...,
             [164, 214, 103],
             [172, 222, 112],
             [168, 220, 108]],
    
            [[104, 181,  26],
             [114, 177,  32],
             [ 75, 159,   0],
             ...,
             [170, 220, 110],
             [168, 220, 110],
             [165, 217, 106]]]], dtype=uint8),
     'image/validation_outputs/7': array([[[[161, 172,  20],
             [156, 172,  18],
             [157, 170,  21],
             ...,
             [182, 218,  90],
             [181, 217,  90],
             [179, 217,  87]],
    
            [[156, 173,  17],
             [154, 170,  21],
             [145, 166,  23],
             ...,
             [182, 219,  91],
             [182, 218,  90],
             [182, 219,  89]],
    
            [[156, 170,  16],
             [158, 170,  21],
             [149, 169,  25],
             ...,
             [182, 220,  91],
             [182, 219,  91],
             [180, 219,  90]],
    
            ...,
    
            [[157, 169,  22],
             [160, 173,  25],
             [161, 175,  26],
             ...,
             [128, 199, 158],
             [125, 194, 155],
             [125, 194, 152]],
    
            [[157, 169,  21],
             [159, 172,  21],
             [160, 173,  23],
             ...,
             [127, 198, 160],
             [124, 196, 158],
             [125, 197, 157]],
    
            [[156, 167,  21],
             [157, 170,  22],
             [160, 173,  24],
             ...,
             [125, 196, 158],
             [124, 197, 158],
             [128, 201, 164]]]], dtype=uint8),
     'image/validation_outputs/8': array([[[[117, 151,  19],
             [119, 152,  19],
             [119, 152,  20],
             ...,
             [116, 150,  22],
             [118, 150,  23],
             [114, 149,  20]],
    
            [[117, 151,  19],
             [119, 152,  22],
             [114, 148,  20],
             ...,
             [119, 152,  25],
             [116, 148,  22],
             [113, 148,  23]],
    
            [[120, 152,  21],
             [120, 153,  23],
             [ 98, 131,   5],
             ...,
             [122, 154,  25],
             [117, 149,  23],
             [114, 148,  22]],
    
            ...,
    
            [[122, 155,  23],
             [111, 144,  17],
             [111, 146,  17],
             ...,
             [126, 158,  33],
             [132, 163,  33],
             [126, 158,  31]],
    
            [[116, 148,  17],
             [108, 142,  16],
             [113, 147,  19],
             ...,
             [127, 158,  33],
             [127, 158,  30],
             [128, 159,  31]],
    
            [[113, 147,  16],
             [108, 143,  14],
             [109, 143,  13],
             ...,
             [122, 155,  30],
             [125, 156,  29],
             [125, 156,  30]]]], dtype=uint8),
     'image/validation_outputs/9': array([[[[160, 209,  91],
             [159, 209,  92],
             [160, 212,  94],
             ...,
             [146, 191,  52],
             [144, 189,  50],
             [143, 188,  49]],
    
            [[159, 209,  90],
             [159, 209,  92],
             [161, 212,  95],
             ...,
             [144, 190,  50],
             [143, 188,  49],
             [144, 189,  50]],
    
            [[160, 210,  91],
             [158, 209,  91],
             [160, 212,  94],
             ...,
             [146, 191,  52],
             [144, 190,  50],
             [143, 188,  49]],
    
            ...,
    
            [[124, 174,  37],
             [124, 174,  37],
             [125, 174,  37],
             ...,
             [117, 159,  20],
             [116, 158,  19],
             [115, 157,  19]],
    
            [[124, 174,  35],
             [124, 174,  37],
             [126, 175,  38],
             ...,
             [119, 161,  22],
             [117, 159,  20],
             [116, 158,  19]],
    
            [[125, 175,  36],
             [126, 176,  39],
             [128, 178,  41],
             ...,
             [118, 159,  20],
             [117, 158,  19],
             [116, 157,  18]]]], dtype=uint8),
     'iou/0': 0.79509777,
     'iou/1': 0.8881304,
     'iou/2': 0.33292255,
     'mean_iou': 0.67205024,
     'steps_per_second': 78.04197055734723,
     'validation_loss': 0.38547286}
train | step:   1032 | training until step 1376...
train | step:   1376 | steps/sec:    5.9 | output: 
    {'learning_rate': 0.022156214,
     'mean_iou': 0.69301605,
     'training_loss': 0.3454611}
saved checkpoint to ./trained_model/ckpt-1376.
 eval | step:   1376 | running 344 steps of evaluation...
 eval | step:   1376 | steps/sec:  106.0 | eval time:    3.2 sec | output: 
    {'best_mean_iou': 0.6818839,
     'image/validation_outputs/0': array([[[[120, 173,  62],
             [129, 175,  65],
             [129, 175,  67],
             ...,
             [136, 185,  80],
             [137, 186,  81],
             [137, 185,  78]],
    
            [[122, 175,  60],
             [128, 179,  65],
             [127, 179,  64],
             ...,
             [133, 185,  79],
             [138, 189,  85],
             [143, 189,  85]],
    
            [[105, 152,  30],
             [108, 155,  33],
             [105, 149,  26],
             ...,
             [144, 193,  88],
             [143, 196,  89],
             [141, 194,  88]],
    
            ...,
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [104, 164,  58],
             [101, 163,  56],
             [101, 160,  53]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [104, 165,  57],
             [101, 163,  55],
             [ 93, 152,  44]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [100, 164,  55],
             [ 89, 149,  41],
             [ 59, 110,   5]]]], dtype=uint8),
     'image/validation_outputs/1': array([[[[129, 190, 105],
             [130, 190, 106],
             [131, 191, 106],
             ...,
             [169, 228, 137],
             [168, 227, 137],
             [165, 226, 135]],
    
            [[131, 191, 107],
             [132, 192, 107],
             [132, 193, 108],
             ...,
             [170, 229, 138],
             [169, 228, 137],
             [168, 228, 137]],
    
            [[131, 191, 107],
             [132, 192, 107],
             [132, 193, 108],
             ...,
             [170, 229, 139],
             [170, 229, 139],
             [167, 227, 136]],
    
            ...,
    
            [[ 81, 124,  23],
             [ 81, 123,  23],
             [ 81, 124,  23],
             ...,
             [ 77, 119,  20],
             [ 77, 118,  20],
             [ 77, 119,  20]],
    
            [[ 80, 122,  22],
             [ 80, 122,  22],
             [ 83, 125,  25],
             ...,
             [ 77, 118,  20],
             [ 77, 118,  20],
             [ 75, 118,  19]],
    
            [[ 77, 121,  19],
             [ 78, 121,  19],
             [ 81, 124,  22],
             ...,
             [ 77, 119,  20],
             [ 77, 118,  20],
             [ 77, 119,  20]]]], dtype=uint8),
     'image/validation_outputs/10': array([[[[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            ...,
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]]]], dtype=uint8),
     'image/validation_outputs/11': array([[[[126, 176,  29],
             [130, 177,  34],
             [128, 176,  34],
             ...,
             [123, 174,  27],
             [121, 172,  26],
             [120, 171,  26]],
    
            [[127, 176,  32],
             [126, 172,  31],
             [111, 157,  17],
             ...,
             [120, 173,  26],
             [120, 172,  26],
             [120, 172,  27]],
    
            [[124, 175,  28],
             [126, 173,  30],
             [126, 175,  33],
             ...,
             [119, 173,  25],
             [117, 172,  24],
             [119, 172,  25]],
    
            ...,
    
            [[100, 161,  17],
             [106, 163,  23],
             [108, 163,  19],
             ...,
             [107, 166,  22],
             [ 98, 158,  13],
             [100, 163,  14]],
    
            [[ 99, 164,  15],
             [100, 164,  19],
             [ 99, 159,  11],
             ...,
             [113, 166,  28],
             [110, 163,  25],
             [105, 164,  20]],
    
            [[106, 167,  20],
             [ 98, 160,  14],
             [ 99, 154,  10],
             ...,
             [109, 164,  24],
             [110, 163,  28],
             [109, 162,  22]]]], dtype=uint8),
     'image/validation_outputs/12': array([[[[125, 166,  55],
             [123, 166,  55],
             [126, 169,  60],
             ...,
             [119, 167,  54],
             [118, 166,  55],
             [116, 163,  52]],
    
            [[123, 167,  55],
             [123, 167,  56],
             [122, 166,  56],
             ...,
             [117, 165,  52],
             [115, 163,  52],
             [114, 161,  50]],
    
            [[122, 166,  54],
             [122, 166,  55],
             [101, 145,  36],
             ...,
             [115, 163,  50],
             [113, 161,  51],
             [114, 161,  50]],
    
            ...,
    
            [[116, 159,  49],
             [109, 151,  41],
             [116, 160,  51],
             ...,
             [107, 150,  41],
             [111, 155,  46],
             [120, 164,  55]],
    
            [[108, 151,  41],
             [101, 144,  35],
             [112, 156,  47],
             ...,
             [115, 160,  51],
             [117, 161,  53],
             [126, 171,  61]],
    
            [[104, 147,  38],
             [103, 145,  37],
             [107, 151,  42],
             ...,
             [121, 166,  55],
             [117, 161,  52],
             [130, 175,  64]]]], dtype=uint8),
     'image/validation_outputs/2': array([[[[188, 250, 149],
             [192, 249, 149],
             [197, 252, 151],
             ...,
             [ 51, 114,  19],
             [ 53, 115,  21],
             [ 52, 114,  19]],
    
            [[191, 251, 151],
             [186, 250, 151],
             [185, 245, 144],
             ...,
             [ 53, 115,  20],
             [ 53, 115,  22],
             [ 52, 115,  20]],
    
            [[185, 245, 147],
             [177, 245, 146],
             [138, 203, 104],
             ...,
             [ 52, 115,  20],
             [ 53, 115,  22],
             [ 53, 115,  22]],
    
            ...,
    
            [[134, 206, 114],
             [135, 205, 113],
             [135, 206, 114],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[134, 205, 113],
             [135, 204, 113],
             [134, 203, 112],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[132, 202, 110],
             [133, 202, 110],
             [132, 202, 109],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]]]], dtype=uint8),
     'image/validation_outputs/3': array([[[[ 59, 104,   1],
             [ 51, 103,   0],
             [ 66, 112,   7],
             ...,
             [ 68, 120,   8],
             [ 71, 122,  12],
             [ 73, 123,  12]],
    
            [[ 58, 104,   0],
             [ 51, 103,   0],
             [ 65, 112,   6],
             ...,
             [ 68, 120,  10],
             [ 71, 123,  13],
             [ 72, 124,  12]],
    
            [[ 59, 105,   1],
             [ 51, 103,   0],
             [ 60, 107,   1],
             ...,
             [ 66, 118,   8],
             [ 68, 119,  10],
             [ 68, 119,   9]],
    
            ...,
    
            [[ 61, 106,   0],
             [ 60, 106,   0],
             [ 51, 103,   5],
             ...,
             [ 62, 112,   1],
             [ 63, 112,   2],
             [ 63, 110,   2]],
    
            [[ 60, 106,   0],
             [ 60, 107,   0],
             [ 51, 104,   6],
             ...,
             [ 65, 112,   2],
             [ 63, 110,   1],
             [ 65, 111,   1]],
    
            [[ 60, 105,   0],
             [ 59, 106,   0],
             [ 53, 106,   7],
             ...,
             [ 63, 110,   2],
             [ 63, 109,   2],
             [ 63, 110,   0]]]], dtype=uint8),
     'image/validation_outputs/4': array([[[[145, 188,  78],
             [145, 189,  80],
             [144, 188,  79],
             ...,
             [140, 179,  67],
             [143, 182,  72],
             [141, 185,  70]],
    
            [[143, 187,  79],
             [143, 187,  79],
             [140, 184,  76],
             ...,
             [134, 176,  65],
             [140, 184,  71],
             [139, 182,  70]],
    
            [[143, 187,  79],
             [147, 190,  82],
             [120, 163,  56],
             ...,
             [125, 167,  58],
             [125, 169,  57],
             [123, 170,  58]],
    
            ...,
    
            [[137, 175,  64],
             [129, 170,  57],
             [131, 170,  61],
             ...,
             [136, 178,  68],
             [132, 175,  63],
             [118, 160,  49]],
    
            [[ 95, 136,  21],
             [140, 182,  70],
             [140, 181,  73],
             ...,
             [129, 170,  61],
             [128, 170,  59],
             [130, 172,  60]],
    
            [[ 95, 139,  21],
             [138, 184,  72],
             [142, 184,  77],
             ...,
             [150, 190,  82],
             [151, 194,  83],
             [140, 181,  71]]]], dtype=uint8),
     'image/validation_outputs/5': array([[[[ 74, 119,  12],
             [ 71, 118,  10],
             [ 65, 113,   7],
             ...,
             [ 63, 118,   4],
             [ 68, 122,   5],
             [ 60, 115,   2]],
    
            [[ 77, 121,  14],
             [ 78, 121,  13],
             [ 77, 120,  12],
             ...,
             [ 63, 118,   4],
             [ 77, 126,   9],
             [ 54, 107,   0]],
    
            [[ 65, 113,   6],
             [ 83, 127,  19],
             [ 78, 121,  12],
             ...,
             [ 60, 118,   5],
             [ 72, 122,   9],
             [ 62, 112,   5]],
    
            ...,
    
            [[141, 206, 112],
             [141, 206, 113],
             [142, 208, 112],
             ...,
             [149, 211, 115],
             [147, 209, 113],
             [144, 208, 111]],
    
            [[134, 203, 110],
             [135, 203, 112],
             [136, 205, 112],
             ...,
             [140, 204, 110],
             [138, 203, 108],
             [135, 203, 108]],
    
            [[123, 196, 105],
             [124, 196, 107],
             [125, 198, 107],
             ...,
             [140, 202, 109],
             [138, 203, 109],
             [138, 202, 107]]]], dtype=uint8),
     'image/validation_outputs/6': array([[[[147, 212,  98],
             [174, 236, 126],
             [169, 229, 121],
             ...,
             [ 57, 121,  17],
             [132, 195,  94],
             [ 95, 155,  58]],
    
            [[141, 209,  96],
             [153, 217, 109],
             [132, 192,  83],
             ...,
             [ 96, 159,  59],
             [ 86, 149,  50],
             [137, 199,  95]],
    
            [[152, 212, 109],
             [180, 239, 137],
             [158, 218, 111],
             ...,
             [187, 243, 147],
             [189, 245, 146],
             [142, 203,  98]],
    
            ...,
    
            [[ 96, 181,  22],
             [ 93, 181,  11],
             [ 80, 175,   2],
             ...,
             [174, 225, 113],
             [173, 224, 113],
             [167, 218, 105]],
    
            [[102, 189,  27],
             [100, 183,  19],
             [ 85, 170,   5],
             ...,
             [164, 214, 103],
             [172, 222, 112],
             [168, 220, 108]],
    
            [[104, 181,  26],
             [114, 177,  32],
             [ 75, 159,   0],
             ...,
             [170, 220, 110],
             [168, 220, 110],
             [165, 217, 106]]]], dtype=uint8),
     'image/validation_outputs/7': array([[[[161, 172,  20],
             [156, 172,  18],
             [157, 170,  21],
             ...,
             [182, 218,  90],
             [181, 217,  90],
             [179, 217,  87]],
    
            [[156, 173,  17],
             [154, 170,  21],
             [145, 166,  23],
             ...,
             [182, 219,  91],
             [182, 218,  90],
             [182, 219,  89]],
    
            [[156, 170,  16],
             [158, 170,  21],
             [149, 169,  25],
             ...,
             [182, 220,  91],
             [182, 219,  91],
             [180, 219,  90]],
    
            ...,
    
            [[157, 169,  22],
             [160, 173,  25],
             [161, 175,  26],
             ...,
             [173, 199,  56],
             [170, 194,  53],
             [169, 194,  50]],
    
            [[157, 169,  21],
             [159, 172,  21],
             [160, 173,  23],
             ...,
             [171, 198,  58],
             [168, 196,  56],
             [169, 197,  55]],
    
            [[156, 167,  21],
             [157, 170,  22],
             [160, 173,  24],
             ...,
             [170, 196,  56],
             [169, 197,  56],
             [171, 201,  62]]]], dtype=uint8),
     'image/validation_outputs/8': array([[[[117, 151,  19],
             [119, 152,  19],
             [119, 152,  20],
             ...,
             [116, 150,  22],
             [118, 150,  23],
             [114, 149,  20]],
    
            [[117, 151,  19],
             [119, 152,  22],
             [114, 148,  20],
             ...,
             [119, 152,  25],
             [116, 148,  22],
             [113, 148,  23]],
    
            [[120, 152,  21],
             [120, 153,  23],
             [ 98, 131,   5],
             ...,
             [122, 154,  25],
             [117, 149,  23],
             [114, 148,  22]],
    
            ...,
    
            [[122, 155,  23],
             [111, 144,  17],
             [111, 146,  17],
             ...,
             [126, 158,  33],
             [132, 163,  33],
             [126, 158,  31]],
    
            [[116, 148,  17],
             [108, 142,  16],
             [113, 147,  19],
             ...,
             [127, 158,  33],
             [127, 158,  30],
             [128, 159,  31]],
    
            [[113, 147,  16],
             [108, 143,  14],
             [109, 143,  13],
             ...,
             [122, 155,  30],
             [125, 156,  29],
             [125, 156,  30]]]], dtype=uint8),
     'image/validation_outputs/9': array([[[[160, 209,  91],
             [159, 209,  92],
             [160, 212,  94],
             ...,
             [146, 191,  52],
             [144, 189,  50],
             [143, 188,  49]],
    
            [[159, 209,  90],
             [159, 209,  92],
             [161, 212,  95],
             ...,
             [144, 190,  50],
             [143, 188,  49],
             [144, 189,  50]],
    
            [[160, 210,  91],
             [158, 209,  91],
             [160, 212,  94],
             ...,
             [146, 191,  52],
             [144, 190,  50],
             [143, 188,  49]],
    
            ...,
    
            [[124, 174,  37],
             [124, 174,  37],
             [125, 174,  37],
             ...,
             [111, 159,  20],
             [109, 158,  19],
             [109, 157,  19]],
    
            [[124, 174,  35],
             [124, 174,  37],
             [126, 175,  38],
             ...,
             [112, 161,  22],
             [111, 159,  20],
             [109, 158,  19]],
    
            [[125, 175,  36],
             [126, 176,  39],
             [128, 178,  41],
             ...,
             [111, 159,  20],
             [110, 158,  19],
             [109, 157,  18]]]], dtype=uint8),
     'iou/0': 0.80969125,
     'iou/1': 0.8898738,
     'iou/2': 0.34608668,
     'mean_iou': 0.6818839,
     'steps_per_second': 105.96190634120993,
     'validation_loss': 0.36764288}
train | step:   1376 | training until step 1720...
train | step:   1720 | steps/sec:    5.9 | output: 
    {'learning_rate': 0.0047586444,
     'mean_iou': 0.70820075,
     'training_loss': 0.32527226}
saved checkpoint to ./trained_model/ckpt-1720.
 eval | step:   1720 | running 344 steps of evaluation...
 eval | step:   1720 | steps/sec:  106.9 | eval time:    3.2 sec | output: 
    {'best_mean_iou': 0.7017154,
     'image/validation_outputs/0': array([[[[120, 173,  62],
             [129, 175,  65],
             [129, 175,  67],
             ...,
             [136, 185,  80],
             [137, 186,  81],
             [137, 185,  78]],
    
            [[122, 175,  60],
             [128, 179,  65],
             [127, 179,  64],
             ...,
             [133, 185,  79],
             [138, 189,  85],
             [143, 189,  85]],
    
            [[105, 152,  30],
             [108, 155,  33],
             [105, 149,  26],
             ...,
             [144, 193,  88],
             [143, 196,  89],
             [141, 194,  88]],
    
            ...,
    
            [[  0, 102, 102],
             [  0, 102, 102],
             [  0, 102, 102],
             ...,
             [104, 164,  58],
             [101, 163,  56],
             [101, 160,  53]],
    
            [[  0, 102, 102],
             [  0, 102, 102],
             [  0, 102, 102],
             ...,
             [104, 165,  57],
             [101, 163,  55],
             [ 93, 152,  44]],
    
            [[  0, 102, 102],
             [  0, 102, 102],
             [  0, 102, 102],
             ...,
             [100, 164,  55],
             [ 89, 149,  41],
             [ 59, 110,   5]]]], dtype=uint8),
     'image/validation_outputs/1': array([[[[129, 190, 105],
             [130, 190, 106],
             [131, 191, 106],
             ...,
             [169, 228, 137],
             [168, 227, 137],
             [165, 226, 135]],
    
            [[131, 191, 107],
             [132, 192, 107],
             [132, 193, 108],
             ...,
             [170, 229, 138],
             [169, 228, 137],
             [168, 228, 137]],
    
            [[131, 191, 107],
             [132, 192, 107],
             [132, 193, 108],
             ...,
             [170, 229, 139],
             [170, 229, 139],
             [167, 227, 136]],
    
            ...,
    
            [[ 81, 124,  23],
             [ 81, 123,  23],
             [ 81, 124,  23],
             ...,
             [ 77, 119,  20],
             [ 77, 118,  20],
             [ 77, 119,  20]],
    
            [[ 80, 122,  22],
             [ 80, 122,  22],
             [ 83, 125,  25],
             ...,
             [ 77, 118,  20],
             [ 77, 118,  20],
             [ 75, 118,  19]],
    
            [[ 77, 121,  19],
             [ 78, 121,  19],
             [ 81, 124,  22],
             ...,
             [ 77, 119,  20],
             [ 77, 118,  20],
             [ 77, 119,  20]]]], dtype=uint8),
     'image/validation_outputs/10': array([[[[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            ...,
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]]]], dtype=uint8),
     'image/validation_outputs/11': array([[[[126, 176,  29],
             [130, 177,  34],
             [128, 176,  34],
             ...,
             [123, 174,  27],
             [121, 172,  26],
             [120, 171,  26]],
    
            [[127, 176,  32],
             [126, 172,  31],
             [111, 157,  17],
             ...,
             [120, 173,  26],
             [120, 172,  26],
             [120, 172,  27]],
    
            [[124, 175,  28],
             [126, 173,  30],
             [126, 175,  33],
             ...,
             [119, 173,  25],
             [117, 172,  24],
             [119, 172,  25]],
    
            ...,
    
            [[100, 161,  17],
             [106, 163,  23],
             [108, 163,  19],
             ...,
             [107, 166,  22],
             [ 98, 158,  13],
             [100, 163,  14]],
    
            [[ 99, 164,  15],
             [100, 164,  19],
             [ 99, 159,  11],
             ...,
             [113, 166,  28],
             [110, 163,  25],
             [105, 164,  20]],
    
            [[106, 167,  20],
             [ 98, 160,  14],
             [ 99, 154,  10],
             ...,
             [109, 164,  24],
             [110, 163,  28],
             [109, 162,  22]]]], dtype=uint8),
     'image/validation_outputs/12': array([[[[125, 166,  55],
             [123, 166,  55],
             [126, 169,  60],
             ...,
             [119, 167,  54],
             [118, 166,  55],
             [116, 163,  52]],
    
            [[123, 167,  55],
             [123, 167,  56],
             [122, 166,  56],
             ...,
             [117, 165,  52],
             [115, 163,  52],
             [114, 161,  50]],
    
            [[122, 166,  54],
             [122, 166,  55],
             [101, 145,  36],
             ...,
             [115, 163,  50],
             [113, 161,  51],
             [114, 161,  50]],
    
            ...,
    
            [[116, 159,  49],
             [109, 151,  41],
             [116, 160,  51],
             ...,
             [107, 150,  41],
             [111, 155,  46],
             [120, 164,  55]],
    
            [[108, 151,  41],
             [101, 144,  35],
             [112, 156,  47],
             ...,
             [115, 160,  51],
             [117, 161,  53],
             [126, 171,  61]],
    
            [[104, 147,  38],
             [103, 145,  37],
             [107, 151,  42],
             ...,
             [121, 166,  55],
             [117, 161,  52],
             [130, 175,  64]]]], dtype=uint8),
     'image/validation_outputs/2': array([[[[188, 250, 149],
             [192, 249, 149],
             [197, 252, 151],
             ...,
             [ 51, 114,  19],
             [ 53, 115,  21],
             [ 52, 114,  19]],
    
            [[191, 251, 151],
             [186, 250, 151],
             [185, 245, 144],
             ...,
             [ 53, 115,  20],
             [ 53, 115,  22],
             [ 52, 115,  20]],
    
            [[185, 245, 147],
             [177, 245, 146],
             [138, 203, 104],
             ...,
             [ 52, 115,  20],
             [ 53, 115,  22],
             [ 53, 115,  22]],
    
            ...,
    
            [[134, 206, 114],
             [135, 205, 113],
             [135, 206, 114],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[134, 205, 113],
             [135, 204, 113],
             [134, 203, 112],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[132, 202, 110],
             [133, 202, 110],
             [132, 202, 109],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]]]], dtype=uint8),
     'image/validation_outputs/3': array([[[[ 59, 104,   1],
             [ 51, 103,   0],
             [ 66, 112,   7],
             ...,
             [ 68, 120,   8],
             [ 71, 122,  12],
             [ 73, 123,  12]],
    
            [[ 58, 104,   0],
             [ 51, 103,   0],
             [ 65, 112,   6],
             ...,
             [ 68, 120,  10],
             [ 71, 123,  13],
             [ 72, 124,  12]],
    
            [[ 59, 105,   1],
             [ 51, 103,   0],
             [ 60, 107,   1],
             ...,
             [ 66, 118,   8],
             [ 68, 119,  10],
             [ 68, 119,   9]],
    
            ...,
    
            [[ 61, 106,   0],
             [ 60, 106,   0],
             [ 51, 103,   5],
             ...,
             [ 62, 112,   1],
             [ 63, 112,   2],
             [ 63, 110,   2]],
    
            [[ 60, 106,   0],
             [ 60, 107,   0],
             [ 51, 104,   6],
             ...,
             [ 65, 112,   2],
             [ 63, 110,   1],
             [ 65, 111,   1]],
    
            [[ 60, 105,   0],
             [ 59, 106,   0],
             [ 53, 106,   7],
             ...,
             [ 63, 110,   2],
             [ 63, 109,   2],
             [ 63, 110,   0]]]], dtype=uint8),
     'image/validation_outputs/4': array([[[[145, 188,  78],
             [145, 189,  80],
             [144, 188,  79],
             ...,
             [140, 179,  67],
             [143, 182,  72],
             [141, 185,  70]],
    
            [[143, 187,  79],
             [143, 187,  79],
             [140, 184,  76],
             ...,
             [134, 176,  65],
             [140, 184,  71],
             [139, 182,  70]],
    
            [[143, 187,  79],
             [147, 190,  82],
             [120, 163,  56],
             ...,
             [125, 167,  58],
             [125, 169,  57],
             [123, 170,  58]],
    
            ...,
    
            [[137, 175,  64],
             [129, 170,  57],
             [131, 170,  61],
             ...,
             [136, 178,  68],
             [132, 175,  63],
             [118, 160,  49]],
    
            [[ 95, 136,  21],
             [140, 182,  70],
             [140, 181,  73],
             ...,
             [129, 170,  61],
             [128, 170,  59],
             [130, 172,  60]],
    
            [[ 95, 139,  21],
             [138, 184,  72],
             [142, 184,  77],
             ...,
             [150, 190,  82],
             [151, 194,  83],
             [140, 181,  71]]]], dtype=uint8),
     'image/validation_outputs/5': array([[[[ 74, 119,  12],
             [ 71, 118,  10],
             [ 65, 113,   7],
             ...,
             [ 63, 118,   4],
             [ 68, 122,   5],
             [ 60, 115,   2]],
    
            [[ 77, 121,  14],
             [ 78, 121,  13],
             [ 77, 120,  12],
             ...,
             [ 63, 118,   4],
             [ 77, 126,   9],
             [ 54, 107,   0]],
    
            [[ 65, 113,   6],
             [ 83, 127,  19],
             [ 78, 121,  12],
             ...,
             [ 60, 118,   5],
             [ 72, 122,   9],
             [ 62, 112,   5]],
    
            ...,
    
            [[141, 206, 112],
             [141, 206, 113],
             [142, 208, 112],
             ...,
             [149, 211, 115],
             [147, 209, 113],
             [144, 208, 111]],
    
            [[134, 203, 110],
             [135, 203, 112],
             [136, 205, 112],
             ...,
             [140, 204, 110],
             [138, 203, 108],
             [135, 203, 108]],
    
            [[123, 196, 105],
             [124, 196, 107],
             [125, 198, 107],
             ...,
             [140, 202, 109],
             [138, 203, 109],
             [138, 202, 107]]]], dtype=uint8),
     'image/validation_outputs/6': array([[[[147, 212,  98],
             [174, 236, 126],
             [169, 229, 121],
             ...,
             [ 57, 121,  17],
             [132, 195,  94],
             [ 95, 155,  58]],
    
            [[141, 209,  96],
             [153, 217, 109],
             [132, 192,  83],
             ...,
             [ 96, 159,  59],
             [ 86, 149,  50],
             [137, 199,  95]],
    
            [[152, 212, 109],
             [180, 239, 137],
             [158, 218, 111],
             ...,
             [187, 243, 147],
             [189, 245, 146],
             [142, 203,  98]],
    
            ...,
    
            [[ 96, 181,  22],
             [ 93, 181,  11],
             [ 80, 175,   2],
             ...,
             [174, 225, 113],
             [173, 224, 113],
             [167, 218, 105]],
    
            [[102, 189,  27],
             [100, 183,  19],
             [ 85, 170,   5],
             ...,
             [164, 214, 103],
             [172, 222, 112],
             [168, 220, 108]],
    
            [[104, 181,  26],
             [114, 177,  32],
             [ 75, 159,   0],
             ...,
             [170, 220, 110],
             [168, 220, 110],
             [165, 217, 106]]]], dtype=uint8),
     'image/validation_outputs/7': array([[[[161, 172,  20],
             [156, 172,  18],
             [157, 170,  21],
             ...,
             [182, 218,  90],
             [181, 217,  90],
             [179, 217,  87]],
    
            [[156, 173,  17],
             [154, 170,  21],
             [145, 166,  23],
             ...,
             [182, 219,  91],
             [182, 218,  90],
             [182, 219,  89]],
    
            [[156, 170,  16],
             [158, 170,  21],
             [149, 169,  25],
             ...,
             [182, 220,  91],
             [182, 219,  91],
             [180, 219,  90]],
    
            ...,
    
            [[157, 169,  22],
             [160, 173,  25],
             [161, 175,  26],
             ...,
             [174, 199,  56],
             [171, 194,  53],
             [170, 194,  50]],
    
            [[157, 169,  21],
             [159, 172,  21],
             [160, 173,  23],
             ...,
             [173, 198,  58],
             [169, 196,  56],
             [171, 197,  55]],
    
            [[156, 167,  21],
             [157, 170,  22],
             [160, 173,  24],
             ...,
             [171, 196,  56],
             [170, 197,  56],
             [173, 201,  62]]]], dtype=uint8),
     'image/validation_outputs/8': array([[[[117, 151,  19],
             [119, 152,  19],
             [119, 152,  20],
             ...,
             [116, 150,  22],
             [118, 150,  23],
             [114, 149,  20]],
    
            [[117, 151,  19],
             [119, 152,  22],
             [114, 148,  20],
             ...,
             [119, 152,  25],
             [116, 148,  22],
             [113, 148,  23]],
    
            [[120, 152,  21],
             [120, 153,  23],
             [ 98, 131,   5],
             ...,
             [122, 154,  25],
             [117, 149,  23],
             [114, 148,  22]],
    
            ...,
    
            [[122, 155,  23],
             [111, 144,  17],
             [111, 146,  17],
             ...,
             [126, 158,  33],
             [132, 163,  33],
             [126, 158,  31]],
    
            [[116, 148,  17],
             [108, 142,  16],
             [113, 147,  19],
             ...,
             [127, 158,  33],
             [127, 158,  30],
             [128, 159,  31]],
    
            [[113, 147,  16],
             [108, 143,  14],
             [109, 143,  13],
             ...,
             [122, 155,  30],
             [125, 156,  29],
             [125, 156,  30]]]], dtype=uint8),
     'image/validation_outputs/9': array([[[[160, 209,  91],
             [159, 209,  92],
             [160, 212,  94],
             ...,
             [146, 191,  52],
             [144, 189,  50],
             [143, 188,  49]],
    
            [[159, 209,  90],
             [159, 209,  92],
             [161, 212,  95],
             ...,
             [144, 190,  50],
             [143, 188,  49],
             [144, 189,  50]],
    
            [[160, 210,  91],
             [158, 209,  91],
             [160, 212,  94],
             ...,
             [146, 191,  52],
             [144, 190,  50],
             [143, 188,  49]],
    
            ...,
    
            [[124, 174,  37],
             [124, 174,  37],
             [125, 174,  37],
             ...,
             [111, 159,  20],
             [109, 158,  19],
             [109, 157,  19]],
    
            [[124, 174,  35],
             [124, 174,  37],
             [126, 175,  38],
             ...,
             [112, 161,  22],
             [111, 159,  20],
             [109, 158,  19]],
    
            [[125, 175,  36],
             [126, 176,  39],
             [128, 178,  41],
             ...,
             [111, 159,  20],
             [110, 158,  19],
             [109, 157,  18]]]], dtype=uint8),
     'iou/0': 0.81555814,
     'iou/1': 0.90009934,
     'iou/2': 0.38948876,
     'mean_iou': 0.7017154,
     'steps_per_second': 106.8744672167273,
     'validation_loss': 0.3554114}
train | step:   1720 | training until step 2000...
train | step:   2000 | steps/sec:    5.9 | output: {'learning_rate': 0.0, 'mean_iou': 0.7126561, 'training_loss': 0.31940305}
 eval | step:   2000 | running 344 steps of evaluation...
 eval | step:   2000 | steps/sec:  107.9 | eval time:    3.2 sec | output: 
    {'best_mean_iou': 0.7045245,
     'image/validation_outputs/0': array([[[[120, 173,  62],
             [129, 175,  65],
             [129, 175,  67],
             ...,
             [136, 185,  80],
             [137, 186,  81],
             [137, 185,  78]],
    
            [[122, 175,  60],
             [128, 179,  65],
             [127, 179,  64],
             ...,
             [133, 185,  79],
             [138, 189,  85],
             [143, 189,  85]],
    
            [[105, 152,  30],
             [108, 155,  33],
             [105, 149,  26],
             ...,
             [144, 193,  88],
             [143, 196,  89],
             [141, 194,  88]],
    
            ...,
    
            [[  0, 102, 102],
             [  0, 102, 102],
             [  0, 102, 102],
             ...,
             [104, 164,  58],
             [101, 163,  56],
             [101, 160,  53]],
    
            [[  0, 102, 102],
             [  0, 102, 102],
             [  0, 102, 102],
             ...,
             [104, 165,  57],
             [101, 163,  55],
             [ 93, 152,  44]],
    
            [[  0, 102, 102],
             [  0, 102, 102],
             [  0, 102, 102],
             ...,
             [100, 164,  55],
             [ 89, 149,  41],
             [ 59, 110,   5]]]], dtype=uint8),
     'image/validation_outputs/1': array([[[[129, 190, 105],
             [130, 190, 106],
             [131, 191, 106],
             ...,
             [169, 228, 137],
             [168, 227, 137],
             [165, 226, 135]],
    
            [[131, 191, 107],
             [132, 192, 107],
             [132, 193, 108],
             ...,
             [170, 229, 138],
             [169, 228, 137],
             [168, 228, 137]],
    
            [[131, 191, 107],
             [132, 192, 107],
             [132, 193, 108],
             ...,
             [170, 229, 139],
             [170, 229, 139],
             [167, 227, 136]],
    
            ...,
    
            [[ 81, 124,  23],
             [ 81, 123,  23],
             [ 81, 124,  23],
             ...,
             [ 77, 119,  20],
             [ 77, 118,  20],
             [ 77, 119,  20]],
    
            [[ 80, 122,  22],
             [ 80, 122,  22],
             [ 83, 125,  25],
             ...,
             [ 77, 118,  20],
             [ 77, 118,  20],
             [ 75, 118,  19]],
    
            [[ 77, 121,  19],
             [ 78, 121,  19],
             [ 81, 124,  22],
             ...,
             [ 77, 119,  20],
             [ 77, 118,  20],
             [ 77, 119,  20]]]], dtype=uint8),
     'image/validation_outputs/10': array([[[[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            ...,
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]]]], dtype=uint8),
     'image/validation_outputs/11': array([[[[126, 176,  29],
             [130, 177,  34],
             [128, 176,  34],
             ...,
             [123, 174,  27],
             [121, 172,  26],
             [120, 171,  26]],
    
            [[127, 176,  32],
             [126, 172,  31],
             [111, 157,  17],
             ...,
             [120, 173,  26],
             [120, 172,  26],
             [120, 172,  27]],
    
            [[124, 175,  28],
             [126, 173,  30],
             [126, 175,  33],
             ...,
             [119, 173,  25],
             [117, 172,  24],
             [119, 172,  25]],
    
            ...,
    
            [[100, 161,  17],
             [106, 163,  23],
             [108, 163,  19],
             ...,
             [107, 166,  22],
             [ 98, 158,  13],
             [100, 163,  14]],
    
            [[ 99, 164,  15],
             [100, 164,  19],
             [ 99, 159,  11],
             ...,
             [113, 166,  28],
             [110, 163,  25],
             [105, 164,  20]],
    
            [[106, 167,  20],
             [ 98, 160,  14],
             [ 99, 154,  10],
             ...,
             [109, 164,  24],
             [110, 163,  28],
             [109, 162,  22]]]], dtype=uint8),
     'image/validation_outputs/12': array([[[[125, 166,  55],
             [123, 166,  55],
             [126, 169,  60],
             ...,
             [119, 167,  54],
             [118, 166,  55],
             [116, 163,  52]],
    
            [[123, 167,  55],
             [123, 167,  56],
             [122, 166,  56],
             ...,
             [117, 165,  52],
             [115, 163,  52],
             [114, 161,  50]],
    
            [[122, 166,  54],
             [122, 166,  55],
             [101, 145,  36],
             ...,
             [115, 163,  50],
             [113, 161,  51],
             [114, 161,  50]],
    
            ...,
    
            [[116, 159,  49],
             [109, 151,  41],
             [116, 160,  51],
             ...,
             [108, 150,  41],
             [112, 155,  46],
             [121, 164,  55]],
    
            [[108, 151,  41],
             [101, 144,  35],
             [112, 156,  47],
             ...,
             [117, 160,  51],
             [119, 161,  53],
             [128, 171,  61]],
    
            [[104, 147,  38],
             [103, 145,  37],
             [107, 151,  42],
             ...,
             [122, 166,  55],
             [119, 161,  52],
             [131, 175,  64]]]], dtype=uint8),
     'image/validation_outputs/2': array([[[[188, 250, 149],
             [192, 249, 149],
             [197, 252, 151],
             ...,
             [ 52, 114,  19],
             [ 53, 115,  21],
             [ 53, 114,  19]],
    
            [[191, 251, 151],
             [186, 250, 151],
             [185, 245, 144],
             ...,
             [ 54, 115,  20],
             [ 54, 115,  22],
             [ 53, 115,  20]],
    
            [[185, 245, 147],
             [177, 245, 146],
             [138, 203, 104],
             ...,
             [ 52, 115,  20],
             [ 54, 115,  22],
             [ 53, 115,  22]],
    
            ...,
    
            [[134, 206, 114],
             [135, 205, 113],
             [135, 206, 114],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[134, 205, 113],
             [135, 204, 113],
             [134, 203, 112],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[132, 202, 110],
             [133, 202, 110],
             [132, 202, 109],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]]]], dtype=uint8),
     'image/validation_outputs/3': array([[[[ 59, 104,   1],
             [ 51, 103,   0],
             [ 66, 112,   7],
             ...,
             [ 68, 120,   8],
             [ 71, 122,  12],
             [ 73, 123,  12]],
    
            [[ 58, 104,   0],
             [ 51, 103,   0],
             [ 65, 112,   6],
             ...,
             [ 68, 120,  10],
             [ 71, 123,  13],
             [ 72, 124,  12]],
    
            [[ 59, 105,   1],
             [ 51, 103,   0],
             [ 60, 107,   1],
             ...,
             [ 66, 118,   8],
             [ 68, 119,  10],
             [ 68, 119,   9]],
    
            ...,
    
            [[ 61, 106,   0],
             [ 60, 106,   0],
             [ 51, 103,   5],
             ...,
             [ 62, 112,   1],
             [ 63, 112,   2],
             [ 63, 110,   2]],
    
            [[ 60, 106,   0],
             [ 60, 107,   0],
             [ 51, 104,   6],
             ...,
             [ 65, 112,   2],
             [ 63, 110,   1],
             [ 65, 111,   1]],
    
            [[ 60, 105,   0],
             [ 59, 106,   0],
             [ 53, 106,   7],
             ...,
             [ 63, 110,   2],
             [ 63, 109,   2],
             [ 63, 110,   0]]]], dtype=uint8),
     'image/validation_outputs/4': array([[[[145, 188,  78],
             [145, 189,  80],
             [144, 188,  79],
             ...,
             [140, 179,  67],
             [143, 182,  72],
             [141, 185,  70]],
    
            [[143, 187,  79],
             [143, 187,  79],
             [140, 184,  76],
             ...,
             [134, 176,  65],
             [140, 184,  71],
             [139, 182,  70]],
    
            [[143, 187,  79],
             [147, 190,  82],
             [120, 163,  56],
             ...,
             [125, 167,  58],
             [125, 169,  57],
             [123, 170,  58]],
    
            ...,
    
            [[137, 175,  64],
             [129, 170,  57],
             [131, 170,  61],
             ...,
             [136, 178,  68],
             [132, 175,  63],
             [118, 160,  49]],
    
            [[ 95, 136,  21],
             [140, 182,  70],
             [140, 181,  73],
             ...,
             [129, 170,  61],
             [128, 170,  59],
             [130, 172,  60]],
    
            [[ 95, 139,  21],
             [138, 184,  72],
             [142, 184,  77],
             ...,
             [150, 190,  82],
             [151, 194,  83],
             [140, 181,  71]]]], dtype=uint8),
     'image/validation_outputs/5': array([[[[ 74, 119,  12],
             [ 71, 118,  10],
             [ 65, 113,   7],
             ...,
             [ 63, 118,   4],
             [ 68, 122,   5],
             [ 60, 115,   2]],
    
            [[ 77, 121,  14],
             [ 78, 121,  13],
             [ 77, 120,  12],
             ...,
             [ 63, 118,   4],
             [ 77, 126,   9],
             [ 54, 107,   0]],
    
            [[ 65, 113,   6],
             [ 83, 127,  19],
             [ 78, 121,  12],
             ...,
             [ 60, 118,   5],
             [ 72, 122,   9],
             [ 62, 112,   5]],
    
            ...,
    
            [[141, 206, 112],
             [141, 206, 113],
             [142, 208, 112],
             ...,
             [149, 211, 115],
             [147, 209, 113],
             [144, 208, 111]],
    
            [[134, 203, 110],
             [135, 203, 112],
             [136, 205, 112],
             ...,
             [140, 204, 110],
             [138, 203, 108],
             [135, 203, 108]],
    
            [[123, 196, 105],
             [124, 196, 107],
             [125, 198, 107],
             ...,
             [140, 202, 109],
             [138, 203, 109],
             [138, 202, 107]]]], dtype=uint8),
     'image/validation_outputs/6': array([[[[147, 212,  98],
             [174, 236, 126],
             [169, 229, 121],
             ...,
             [ 57, 121,  17],
             [132, 195,  94],
             [ 95, 155,  58]],
    
            [[141, 209,  96],
             [153, 217, 109],
             [132, 192,  83],
             ...,
             [ 96, 159,  59],
             [ 86, 149,  50],
             [137, 199,  95]],
    
            [[152, 212, 109],
             [180, 239, 137],
             [158, 218, 111],
             ...,
             [187, 243, 147],
             [189, 245, 146],
             [142, 203,  98]],
    
            ...,
    
            [[ 96, 181,  22],
             [ 93, 181,  11],
             [ 80, 175,   2],
             ...,
             [174, 225, 113],
             [173, 224, 113],
             [167, 218, 105]],
    
            [[102, 189,  27],
             [100, 183,  19],
             [ 85, 170,   5],
             ...,
             [164, 214, 103],
             [172, 222, 112],
             [168, 220, 108]],
    
            [[104, 181,  26],
             [114, 177,  32],
             [ 75, 159,   0],
             ...,
             [170, 220, 110],
             [168, 220, 110],
             [165, 217, 106]]]], dtype=uint8),
     'image/validation_outputs/7': array([[[[161, 172,  20],
             [156, 172,  18],
             [157, 170,  21],
             ...,
             [182, 218,  90],
             [181, 217,  90],
             [179, 217,  87]],
    
            [[156, 173,  17],
             [154, 170,  21],
             [145, 166,  23],
             ...,
             [182, 219,  91],
             [182, 218,  90],
             [182, 219,  89]],
    
            [[156, 170,  16],
             [158, 170,  21],
             [149, 169,  25],
             ...,
             [182, 220,  91],
             [182, 219,  91],
             [180, 219,  90]],
    
            ...,
    
            [[157, 169,  22],
             [160, 173,  25],
             [161, 175,  26],
             ...,
             [174, 199,  56],
             [171, 194,  53],
             [170, 194,  50]],
    
            [[157, 169,  21],
             [159, 172,  21],
             [160, 173,  23],
             ...,
             [173, 198,  58],
             [169, 196,  56],
             [171, 197,  55]],
    
            [[156, 167,  21],
             [157, 170,  22],
             [160, 173,  24],
             ...,
             [171, 196,  56],
             [170, 197,  56],
             [173, 201,  62]]]], dtype=uint8),
     'image/validation_outputs/8': array([[[[117, 151,  19],
             [119, 152,  19],
             [119, 152,  20],
             ...,
             [116, 150,  22],
             [118, 150,  23],
             [114, 149,  20]],
    
            [[117, 151,  19],
             [119, 152,  22],
             [114, 148,  20],
             ...,
             [119, 152,  25],
             [116, 148,  22],
             [113, 148,  23]],
    
            [[120, 152,  21],
             [120, 153,  23],
             [ 98, 131,   5],
             ...,
             [122, 154,  25],
             [117, 149,  23],
             [114, 148,  22]],
    
            ...,
    
            [[122, 155,  23],
             [111, 144,  17],
             [111, 146,  17],
             ...,
             [126, 158,  33],
             [132, 163,  33],
             [126, 158,  31]],
    
            [[116, 148,  17],
             [108, 142,  16],
             [113, 147,  19],
             ...,
             [127, 158,  33],
             [127, 158,  30],
             [128, 159,  31]],
    
            [[113, 147,  16],
             [108, 143,  14],
             [109, 143,  13],
             ...,
             [122, 155,  30],
             [125, 156,  29],
             [125, 156,  30]]]], dtype=uint8),
     'image/validation_outputs/9': array([[[[160, 209,  91],
             [159, 209,  92],
             [160, 212,  94],
             ...,
             [146, 191,  52],
             [144, 189,  50],
             [143, 188,  49]],
    
            [[159, 209,  90],
             [159, 209,  92],
             [161, 212,  95],
             ...,
             [144, 190,  50],
             [143, 188,  49],
             [144, 189,  50]],
    
            [[160, 210,  91],
             [158, 209,  91],
             [160, 212,  94],
             ...,
             [146, 191,  52],
             [144, 190,  50],
             [143, 188,  49]],
    
            ...,
    
            [[124, 174,  37],
             [124, 174,  37],
             [125, 174,  37],
             ...,
             [111, 159,  20],
             [109, 158,  19],
             [109, 157,  19]],
    
            [[124, 174,  35],
             [124, 174,  37],
             [126, 175,  38],
             ...,
             [112, 161,  22],
             [111, 159,  20],
             [109, 158,  19]],
    
            [[125, 175,  36],
             [126, 176,  39],
             [128, 178,  41],
             ...,
             [111, 159,  20],
             [110, 158,  19],
             [109, 157,  18]]]], dtype=uint8),
     'iou/0': 0.8178029,
     'iou/1': 0.9019719,
     'iou/2': 0.3937988,
     'mean_iou': 0.7045245,
     'steps_per_second': 107.90473240334056,
     'validation_loss': 0.353675}
saved checkpoint to ./trained_model/ckpt-2000.
 eval | step:   2000 | running 344 steps of evaluation...
 eval | step:   2000 | steps/sec:  155.2 | eval time:    2.2 sec | output: 
    {'best_mean_iou': 0.7045245,
     'image/validation_outputs/0': array([[[[120, 173,  62],
             [129, 175,  65],
             [129, 175,  67],
             ...,
             [136, 185,  80],
             [137, 186,  81],
             [137, 185,  78]],
    
            [[122, 175,  60],
             [128, 179,  65],
             [127, 179,  64],
             ...,
             [133, 185,  79],
             [138, 189,  85],
             [143, 189,  85]],
    
            [[105, 152,  30],
             [108, 155,  33],
             [105, 149,  26],
             ...,
             [144, 193,  88],
             [143, 196,  89],
             [141, 194,  88]],
    
            ...,
    
            [[  0, 102, 102],
             [  0, 102, 102],
             [  0, 102, 102],
             ...,
             [104, 164,  58],
             [101, 163,  56],
             [101, 160,  53]],
    
            [[  0, 102, 102],
             [  0, 102, 102],
             [  0, 102, 102],
             ...,
             [104, 165,  57],
             [101, 163,  55],
             [ 93, 152,  44]],
    
            [[  0, 102, 102],
             [  0, 102, 102],
             [  0, 102, 102],
             ...,
             [100, 164,  55],
             [ 89, 149,  41],
             [ 59, 110,   5]]]], dtype=uint8),
     'image/validation_outputs/1': array([[[[129, 190, 105],
             [130, 190, 106],
             [131, 191, 106],
             ...,
             [169, 228, 137],
             [168, 227, 137],
             [165, 226, 135]],
    
            [[131, 191, 107],
             [132, 192, 107],
             [132, 193, 108],
             ...,
             [170, 229, 138],
             [169, 228, 137],
             [168, 228, 137]],
    
            [[131, 191, 107],
             [132, 192, 107],
             [132, 193, 108],
             ...,
             [170, 229, 139],
             [170, 229, 139],
             [167, 227, 136]],
    
            ...,
    
            [[ 81, 124,  23],
             [ 81, 123,  23],
             [ 81, 124,  23],
             ...,
             [ 77, 119,  20],
             [ 77, 118,  20],
             [ 77, 119,  20]],
    
            [[ 80, 122,  22],
             [ 80, 122,  22],
             [ 83, 125,  25],
             ...,
             [ 77, 118,  20],
             [ 77, 118,  20],
             [ 75, 118,  19]],
    
            [[ 77, 121,  19],
             [ 78, 121,  19],
             [ 81, 124,  22],
             ...,
             [ 77, 119,  20],
             [ 77, 118,  20],
             [ 77, 119,  20]]]], dtype=uint8),
     'image/validation_outputs/10': array([[[[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            ...,
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]]]], dtype=uint8),
     'image/validation_outputs/11': array([[[[126, 176,  29],
             [130, 177,  34],
             [128, 176,  34],
             ...,
             [123, 174,  27],
             [121, 172,  26],
             [120, 171,  26]],
    
            [[127, 176,  32],
             [126, 172,  31],
             [111, 157,  17],
             ...,
             [120, 173,  26],
             [120, 172,  26],
             [120, 172,  27]],
    
            [[124, 175,  28],
             [126, 173,  30],
             [126, 175,  33],
             ...,
             [119, 173,  25],
             [117, 172,  24],
             [119, 172,  25]],
    
            ...,
    
            [[100, 161,  17],
             [106, 163,  23],
             [108, 163,  19],
             ...,
             [107, 166,  22],
             [ 98, 158,  13],
             [100, 163,  14]],
    
            [[ 99, 164,  15],
             [100, 164,  19],
             [ 99, 159,  11],
             ...,
             [113, 166,  28],
             [110, 163,  25],
             [105, 164,  20]],
    
            [[106, 167,  20],
             [ 98, 160,  14],
             [ 99, 154,  10],
             ...,
             [109, 164,  24],
             [110, 163,  28],
             [109, 162,  22]]]], dtype=uint8),
     'image/validation_outputs/12': array([[[[125, 166,  55],
             [123, 166,  55],
             [126, 169,  60],
             ...,
             [119, 167,  54],
             [118, 166,  55],
             [116, 163,  52]],
    
            [[123, 167,  55],
             [123, 167,  56],
             [122, 166,  56],
             ...,
             [117, 165,  52],
             [115, 163,  52],
             [114, 161,  50]],
    
            [[122, 166,  54],
             [122, 166,  55],
             [101, 145,  36],
             ...,
             [115, 163,  50],
             [113, 161,  51],
             [114, 161,  50]],
    
            ...,
    
            [[116, 159,  49],
             [109, 151,  41],
             [116, 160,  51],
             ...,
             [108, 150,  41],
             [112, 155,  46],
             [121, 164,  55]],
    
            [[108, 151,  41],
             [101, 144,  35],
             [112, 156,  47],
             ...,
             [117, 160,  51],
             [119, 161,  53],
             [128, 171,  61]],
    
            [[104, 147,  38],
             [103, 145,  37],
             [107, 151,  42],
             ...,
             [122, 166,  55],
             [119, 161,  52],
             [131, 175,  64]]]], dtype=uint8),
     'image/validation_outputs/2': array([[[[188, 250, 149],
             [192, 249, 149],
             [197, 252, 151],
             ...,
             [ 52, 114,  19],
             [ 53, 115,  21],
             [ 53, 114,  19]],
    
            [[191, 251, 151],
             [186, 250, 151],
             [185, 245, 144],
             ...,
             [ 54, 115,  20],
             [ 54, 115,  22],
             [ 53, 115,  20]],
    
            [[185, 245, 147],
             [177, 245, 146],
             [138, 203, 104],
             ...,
             [ 52, 115,  20],
             [ 54, 115,  22],
             [ 53, 115,  22]],
    
            ...,
    
            [[134, 206, 114],
             [135, 205, 113],
             [135, 206, 114],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[134, 205, 113],
             [135, 204, 113],
             [134, 203, 112],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]],
    
            [[132, 202, 110],
             [133, 202, 110],
             [132, 202, 109],
             ...,
             [ 51, 102,   0],
             [ 51, 102,   0],
             [ 51, 102,   0]]]], dtype=uint8),
     'image/validation_outputs/3': array([[[[ 59, 104,   1],
             [ 51, 103,   0],
             [ 66, 112,   7],
             ...,
             [ 68, 120,   8],
             [ 71, 122,  12],
             [ 73, 123,  12]],
    
            [[ 58, 104,   0],
             [ 51, 103,   0],
             [ 65, 112,   6],
             ...,
             [ 68, 120,  10],
             [ 71, 123,  13],
             [ 72, 124,  12]],
    
            [[ 59, 105,   1],
             [ 51, 103,   0],
             [ 60, 107,   1],
             ...,
             [ 66, 118,   8],
             [ 68, 119,  10],
             [ 68, 119,   9]],
    
            ...,
    
            [[ 61, 106,   0],
             [ 60, 106,   0],
             [ 51, 103,   5],
             ...,
             [ 62, 112,   1],
             [ 63, 112,   2],
             [ 63, 110,   2]],
    
            [[ 60, 106,   0],
             [ 60, 107,   0],
             [ 51, 104,   6],
             ...,
             [ 65, 112,   2],
             [ 63, 110,   1],
             [ 65, 111,   1]],
    
            [[ 60, 105,   0],
             [ 59, 106,   0],
             [ 53, 106,   7],
             ...,
             [ 63, 110,   2],
             [ 63, 109,   2],
             [ 63, 110,   0]]]], dtype=uint8),
     'image/validation_outputs/4': array([[[[145, 188,  78],
             [145, 189,  80],
             [144, 188,  79],
             ...,
             [140, 179,  67],
             [143, 182,  72],
             [141, 185,  70]],
    
            [[143, 187,  79],
             [143, 187,  79],
             [140, 184,  76],
             ...,
             [134, 176,  65],
             [140, 184,  71],
             [139, 182,  70]],
    
            [[143, 187,  79],
             [147, 190,  82],
             [120, 163,  56],
             ...,
             [125, 167,  58],
             [125, 169,  57],
             [123, 170,  58]],
    
            ...,
    
            [[137, 175,  64],
             [129, 170,  57],
             [131, 170,  61],
             ...,
             [136, 178,  68],
             [132, 175,  63],
             [118, 160,  49]],
    
            [[ 95, 136,  21],
             [140, 182,  70],
             [140, 181,  73],
             ...,
             [129, 170,  61],
             [128, 170,  59],
             [130, 172,  60]],
    
            [[ 95, 139,  21],
             [138, 184,  72],
             [142, 184,  77],
             ...,
             [150, 190,  82],
             [151, 194,  83],
             [140, 181,  71]]]], dtype=uint8),
     'image/validation_outputs/5': array([[[[ 74, 119,  12],
             [ 71, 118,  10],
             [ 65, 113,   7],
             ...,
             [ 63, 118,   4],
             [ 68, 122,   5],
             [ 60, 115,   2]],
    
            [[ 77, 121,  14],
             [ 78, 121,  13],
             [ 77, 120,  12],
             ...,
             [ 63, 118,   4],
             [ 77, 126,   9],
             [ 54, 107,   0]],
    
            [[ 65, 113,   6],
             [ 83, 127,  19],
             [ 78, 121,  12],
             ...,
             [ 60, 118,   5],
             [ 72, 122,   9],
             [ 62, 112,   5]],
    
            ...,
    
            [[141, 206, 112],
             [141, 206, 113],
             [142, 208, 112],
             ...,
             [149, 211, 115],
             [147, 209, 113],
             [144, 208, 111]],
    
            [[134, 203, 110],
             [135, 203, 112],
             [136, 205, 112],
             ...,
             [140, 204, 110],
             [138, 203, 108],
             [135, 203, 108]],
    
            [[123, 196, 105],
             [124, 196, 107],
             [125, 198, 107],
             ...,
             [140, 202, 109],
             [138, 203, 109],
             [138, 202, 107]]]], dtype=uint8),
     'image/validation_outputs/6': array([[[[147, 212,  98],
             [174, 236, 126],
             [169, 229, 121],
             ...,
             [ 57, 121,  17],
             [132, 195,  94],
             [ 95, 155,  58]],
    
            [[141, 209,  96],
             [153, 217, 109],
             [132, 192,  83],
             ...,
             [ 96, 159,  59],
             [ 86, 149,  50],
             [137, 199,  95]],
    
            [[152, 212, 109],
             [180, 239, 137],
             [158, 218, 111],
             ...,
             [187, 243, 147],
             [189, 245, 146],
             [142, 203,  98]],
    
            ...,
    
            [[ 96, 181,  22],
             [ 93, 181,  11],
             [ 80, 175,   2],
             ...,
             [174, 225, 113],
             [173, 224, 113],
             [167, 218, 105]],
    
            [[102, 189,  27],
             [100, 183,  19],
             [ 85, 170,   5],
             ...,
             [164, 214, 103],
             [172, 222, 112],
             [168, 220, 108]],
    
            [[104, 181,  26],
             [114, 177,  32],
             [ 75, 159,   0],
             ...,
             [170, 220, 110],
             [168, 220, 110],
             [165, 217, 106]]]], dtype=uint8),
     'image/validation_outputs/7': array([[[[161, 172,  20],
             [156, 172,  18],
             [157, 170,  21],
             ...,
             [182, 218,  90],
             [181, 217,  90],
             [179, 217,  87]],
    
            [[156, 173,  17],
             [154, 170,  21],
             [145, 166,  23],
             ...,
             [182, 219,  91],
             [182, 218,  90],
             [182, 219,  89]],
    
            [[156, 170,  16],
             [158, 170,  21],
             [149, 169,  25],
             ...,
             [182, 220,  91],
             [182, 219,  91],
             [180, 219,  90]],
    
            ...,
    
            [[157, 169,  22],
             [160, 173,  25],
             [161, 175,  26],
             ...,
             [174, 199,  56],
             [171, 194,  53],
             [170, 194,  50]],
    
            [[157, 169,  21],
             [159, 172,  21],
             [160, 173,  23],
             ...,
             [173, 198,  58],
             [169, 196,  56],
             [171, 197,  55]],
    
            [[156, 167,  21],
             [157, 170,  22],
             [160, 173,  24],
             ...,
             [171, 196,  56],
             [170, 197,  56],
             [173, 201,  62]]]], dtype=uint8),
     'image/validation_outputs/8': array([[[[117, 151,  19],
             [119, 152,  19],
             [119, 152,  20],
             ...,
             [116, 150,  22],
             [118, 150,  23],
             [114, 149,  20]],
    
            [[117, 151,  19],
             [119, 152,  22],
             [114, 148,  20],
             ...,
             [119, 152,  25],
             [116, 148,  22],
             [113, 148,  23]],
    
            [[120, 152,  21],
             [120, 153,  23],
             [ 98, 131,   5],
             ...,
             [122, 154,  25],
             [117, 149,  23],
             [114, 148,  22]],
    
            ...,
    
            [[122, 155,  23],
             [111, 144,  17],
             [111, 146,  17],
             ...,
             [126, 158,  33],
             [132, 163,  33],
             [126, 158,  31]],
    
            [[116, 148,  17],
             [108, 142,  16],
             [113, 147,  19],
             ...,
             [127, 158,  33],
             [127, 158,  30],
             [128, 159,  31]],
    
            [[113, 147,  16],
             [108, 143,  14],
             [109, 143,  13],
             ...,
             [122, 155,  30],
             [125, 156,  29],
             [125, 156,  30]]]], dtype=uint8),
     'image/validation_outputs/9': array([[[[160, 209,  91],
             [159, 209,  92],
             [160, 212,  94],
             ...,
             [146, 191,  52],
             [144, 189,  50],
             [143, 188,  49]],
    
            [[159, 209,  90],
             [159, 209,  92],
             [161, 212,  95],
             ...,
             [144, 190,  50],
             [143, 188,  49],
             [144, 189,  50]],
    
            [[160, 210,  91],
             [158, 209,  91],
             [160, 212,  94],
             ...,
             [146, 191,  52],
             [144, 190,  50],
             [143, 188,  49]],
    
            ...,
    
            [[124, 174,  37],
             [124, 174,  37],
             [125, 174,  37],
             ...,
             [111, 159,  20],
             [109, 158,  19],
             [109, 157,  19]],
    
            [[124, 174,  35],
             [124, 174,  37],
             [126, 175,  38],
             ...,
             [112, 161,  22],
             [111, 159,  20],
             [109, 158,  19]],
    
            [[125, 175,  36],
             [126, 176,  39],
             [128, 178,  41],
             ...,
             [111, 159,  20],
             [110, 158,  19],
             [109, 157,  18]]]], dtype=uint8),
     'iou/0': 0.8178029,
     'iou/1': 0.9019719,
     'iou/2': 0.3937988,
     'mean_iou': 0.7045245,
     'steps_per_second': 155.1795389034424,
     'validation_loss': 0.353675}

Load logs in tensorboard

%load_ext tensorboard
%tensorboard --logdir './trained_model'

Saving and exporting the trained model

The keras.Model object returned by train_lib.run_experiment expects the data to be normalized by the dataset loader using the same mean and variance statiscics in preprocess_ops.normalize_image(image, offset=MEAN_RGB, scale=STDDEV_RGB). This export function handles those details, so you can pass tf.uint8 images and get the correct results.

export_saved_model_lib.export_inference_graph(
    input_type='image_tensor',
    batch_size=1,
    input_image_size=[HEIGHT, WIDTH],
    params=exp_config,
    checkpoint_path=tf.train.latest_checkpoint(model_dir),
    export_dir=export_dir)

WARNING:tensorflow:Skipping full serialization of Keras layer <official.vision.modeling.segmentation_model.SegmentationModel object at 0x7f48ebd4e9a0>, because it is not built.
WARNING:tensorflow:Skipping full serialization of Keras layer <official.vision.modeling.segmentation_model.SegmentationModel object at 0x7f48ebd4e9a0>, because it is not built.
WARNING:tensorflow:Skipping full serialization of Keras layer <keras.src.layers.merging.add.Add object at 0x7f48ebfdbf10>, because it is not built.
WARNING:tensorflow:Skipping full serialization of Keras layer <keras.src.layers.merging.add.Add object at 0x7f48ebfdbf10>, because it is not built.
WARNING:tensorflow:Skipping full serialization of Keras layer <keras.src.layers.merging.add.Add object at 0x7f445f7bee50>, because it is not built.
WARNING:tensorflow:Skipping full serialization of Keras layer <keras.src.layers.merging.add.Add object at 0x7f445f7bee50>, because it is not built.
WARNING:tensorflow:Skipping full serialization of Keras layer <keras.src.layers.merging.add.Add object at 0x7f48d83cce50>, because it is not built.
WARNING:tensorflow:Skipping full serialization of Keras layer <keras.src.layers.merging.add.Add object at 0x7f48d83cce50>, because it is not built.
WARNING:tensorflow:Skipping full serialization of Keras layer <keras.src.layers.merging.add.Add object at 0x7f48d81a9910>, because it is not built.
WARNING:tensorflow:Skipping full serialization of Keras layer <keras.src.layers.merging.add.Add object at 0x7f48d81a9910>, because it is not built.
WARNING:tensorflow:Skipping full serialization of Keras layer <keras.src.layers.merging.add.Add object at 0x7f48ebe584c0>, because it is not built.
WARNING:tensorflow:Skipping full serialization of Keras layer <keras.src.layers.merging.add.Add object at 0x7f48ebe584c0>, because it is not built.
WARNING:tensorflow:Skipping full serialization of Keras layer <keras.src.layers.merging.add.Add object at 0x7f48ebe365b0>, because it is not built.
WARNING:tensorflow:Skipping full serialization of Keras layer <keras.src.layers.merging.add.Add object at 0x7f48ebe365b0>, because it is not built.
WARNING:tensorflow:Skipping full serialization of Keras layer <keras.src.layers.merging.add.Add object at 0x7f48ebdf2130>, because it is not built.
WARNING:tensorflow:Skipping full serialization of Keras layer <keras.src.layers.merging.add.Add object at 0x7f48ebdf2130>, because it is not built.
INFO:tensorflow:Assets written to: ./exported_model/assets
INFO:tensorflow:Assets written to: ./exported_model/assets

Importing SavedModel

imported = tf.saved_model.load(export_dir)
model_fn = imported.signatures['serving_default']

WARNING:absl:Importing a function (__inference_internal_grad_fn_352868) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_358079) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355190) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357917) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352427) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355172) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353219) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353714) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352535) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353471) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353840) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354407) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357728) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357890) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355667) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356054) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353228) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355973) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357773) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357377) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355127) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357944) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353795) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356963) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355019) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357539) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356576) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355406) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353525) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354875) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353066) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356360) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355712) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354956) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353633) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353237) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352364) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357854) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352931) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356099) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354704) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357242) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353768) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355433) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354119) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356072) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356801) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354110) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355487) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352859) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353345) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356504) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354929) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354659) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354812) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354785) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352706) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356234) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354857) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357701) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354803) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353084) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354686) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_358070) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354677) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352805) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352355) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356351) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354767) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353129) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356936) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353318) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356792) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356081) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352382) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355865) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356486) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356630) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353849) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357899) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352481) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356117) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355856) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356027) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354074) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353759) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355757) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354380) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352679) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355991) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355109) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355073) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356009) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353444) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356153) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354479) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352778) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357611) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354848) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356594) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353876) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354083) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357107) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355208) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354209) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354245) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352841) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357476) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356747) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353372) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354578) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355604) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356927) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354299) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356621) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354371) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357404) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356720) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355703) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355163) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356972) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353426) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354695) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357935) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352373) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352886) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353903) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355577) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355037) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353786) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354920) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357512) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353057) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354533) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357737) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356765) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354200) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352670) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356000) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357620) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354911) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354191) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352472) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353075) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353570) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354740) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356882) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354164) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353201) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357602) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356477) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353741) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354308) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354623) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354353) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352877) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357143) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356810) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354137) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353012) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353894) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356837) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355442) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355397) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355118) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352904) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357971) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355658) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355730) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355595) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354461) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355937) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356864) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_358097) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357395) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353183) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357125) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353021) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353993) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356657) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354749) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354542) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353615) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352409) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356432) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354344) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356531) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353210) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356990) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354497) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355910) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356873) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352697) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356693) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352913) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352607) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352328) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354029) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353534) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354650) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355541) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355514) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356261) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355001) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355640) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357323) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352823) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352499) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352742) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357746) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352940) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352850) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355370) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355676) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357278) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353417) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356378) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354281) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354551) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355892) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_358007) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356855) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356999) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355253) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357584) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352571) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354605) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354587) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355469) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355649) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355235) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353390) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353831) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357863) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357413) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356513) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354425) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352949) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357269) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356135) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355784) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353957) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357800) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357521) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355685) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353597) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357179) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357719) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357098) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355244) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354065) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356522) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355100) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353156) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356225) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354506) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354515) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357035) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356342) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354092) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356459) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357314) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353975) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355280) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352895) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355748) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356414) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356279) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352994) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354011) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353255) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353102) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352454) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354632) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356387) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352661) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352724) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353363) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356423) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352985) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355811) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357566) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356846) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354038) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354416) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356774) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353327) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356315) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355964) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353003) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356369) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352445) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355325) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353048) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352625) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_358034) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357062) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354722) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354434) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356045) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357017) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353543) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355847) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357980) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357044) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357152) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356468) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355154) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353498) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353039) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354974) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357161) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353489) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357665) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353912) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355352) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354983) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354560) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352643) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356288) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355550) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354839) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355343) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353813) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357467) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357071) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355082) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357332) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354821) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354389) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353462) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353507) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354362) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353867) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356828) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354776) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355271) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356396) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355586) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357080) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357359) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354326) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354128) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354794) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356549) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356108) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354830) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352580) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353399) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353723) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357053) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353030) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356666) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355523) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356558) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353588) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357026) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353678) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354965) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356891) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352958) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352760) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353984) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354290) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356567) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353669) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353435) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357134) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355199) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356171) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356333) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357683) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352526) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352391) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353930) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357710) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355424) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356675) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353939) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356819) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355226) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357458) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353282) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352436) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353291) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356756) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355379) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357872) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355739) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_358025) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355046) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355388) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355289) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355793) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355334) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355064) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353264) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353552) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354893) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356540) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354524) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356162) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353354) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352976) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355829) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355775) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353147) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357386) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354173) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354596) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356450) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352490) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356783) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355091) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352751) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353165) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354488) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357485) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354227) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356243) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353300) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356900) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353822) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352922) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357287) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354668) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357926) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357647) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357674) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356585) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352616) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357755) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352832) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357224) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354263) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357953) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355946) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355766) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355496) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352544) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355838) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352787) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352715) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356090) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357998) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357368) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352688) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357881) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357008) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357548) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357557) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356036) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356207) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352463) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355217) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353732) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357836) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357575) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356270) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354335) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355982) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357629) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357260) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356297) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352634) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357494) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355694) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355919) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353804) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353480) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352589) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352346) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356198) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353705) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355568) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357593) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357791) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357305) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354101) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353579) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357296) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356495) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352967) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356189) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352769) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357764) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353453) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352652) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355532) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357530) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355820) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354731) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356648) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354047) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355559) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357341) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353777) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352733) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355631) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353138) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353309) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353651) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356252) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355901) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355307) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355928) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356126) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354470) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354236) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355613) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353093) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357089) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357116) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355298) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354947) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353606) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353885) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357197) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354902) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354155) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352418) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354020) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352598) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354254) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356603) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357692) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357350) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352517) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355010) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352508) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355802) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357818) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353948) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355262) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357422) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357827) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353192) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356180) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357170) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352814) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_358043) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356639) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357782) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356909) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353696) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356063) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354758) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356918) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356702) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356216) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354218) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353687) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353624) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352562) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356981) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352337) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357215) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353120) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357440) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354938) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352796) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354056) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357638) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355955) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355721) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355478) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354452) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357431) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354398) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354569) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354002) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355136) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356306) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_358088) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353921) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356144) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353561) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_358061) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357251) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353273) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356684) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355028) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355874) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354713) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354146) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352553) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356441) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356945) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354992) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357449) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356711) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357206) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356612) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353111) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356954) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357962) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353642) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356324) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356729) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356405) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353174) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357989) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357845) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357908) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355145) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355451) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353750) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354884) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353858) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354866) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355460) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354443) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357656) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355316) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353660) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354641) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357809) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354317) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357503) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354614) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353966) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355181) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_352400) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353381) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354182) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_358016) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355622) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355361) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355505) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353408) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_358052) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357233) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356018) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355055) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353246) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_356738) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_357188) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_354272) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353336) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_353516) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355883) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
WARNING:absl:Importing a function (__inference_internal_grad_fn_355415) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.

Visualize predictions

def create_mask(pred_mask):
  pred_mask = tf.math.argmax(pred_mask, axis=-1)
  pred_mask = pred_mask[..., tf.newaxis]
  return pred_mask[0]


for record in test_ds.take(15):
  image = tf.image.resize(record['image'], size=[HEIGHT, WIDTH])
  image = tf.cast(image, dtype=tf.uint8)
  mask = tf.image.resize(record['segmentation_mask'], size=[HEIGHT, WIDTH])
  predicted_mask = model_fn(tf.expand_dims(record['image'], axis=0))
  plot_masks([image, mask, create_mask(predicted_mask['logits'])])

png

png

png

png

png

png

png

png

png

png

png

png

png

png

png