A stateful resource that aggregates statistics from one or more iterators.
tf.compat.v1.data.experimental.StatsAggregator()
To record statistics, use one of the custom transformation functions defined
in this module when defining your tf.data.Dataset. All statistics will be
aggregated by the StatsAggregator that is associated with a particular
iterator (see below). For example, to record the latency of producing each
element by iterating over a dataset:
To associate a StatsAggregator with a tf.data.Dataset object, use
the following pattern:
aggregator=tf.data.experimental.StatsAggregator()dataset=...# Apply `StatsOptions` to associate `dataset` with `aggregator`.options=tf.data.Options()options.experimental_stats.aggregator=aggregatordataset=dataset.with_options(options)
To get a protocol buffer summary of the currently aggregated statistics,
use the StatsAggregator.get_summary() tensor. The easiest way to do this
is to add the returned tensor to the tf.GraphKeys.SUMMARIES collection,
so that the summaries will be included with any existing summaries.
Returns a string tf.Tensor that summarizes the aggregated statistics.
The returned tensor will contain a serialized tf.compat.v1.summary.Summary
protocol
buffer, which can be used with the standard TensorBoard logging facilities.
Returns
A scalar string tf.Tensor that summarizes the aggregated statistics.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Missing the information I need","missingTheInformationINeed","thumb-down"],["Too complicated / too many steps","tooComplicatedTooManySteps","thumb-down"],["Out of date","outOfDate","thumb-down"],["Samples / code issue","samplesCodeIssue","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2020-10-01 UTC."],[],[],null,["# tf.compat.v1.data.experimental.StatsAggregator\n\n\u003cbr /\u003e\n\n|----------------------------------------------------------------------------------------------------------------------------------------------------|\n| [View source on GitHub](https://github.com/tensorflow/tensorflow/blob/v2.0.0/tensorflow/python/data/experimental/ops/stats_aggregator.py#L82-L140) |\n\nA stateful resource that aggregates statistics from one or more iterators. \n\n tf.compat.v1.data.experimental.StatsAggregator()\n\nTo record statistics, use one of the custom transformation functions defined\nin this module when defining your [`tf.data.Dataset`](../../../../../tf/data/Dataset). All statistics will be\naggregated by the `StatsAggregator` that is associated with a particular\niterator (see below). For example, to record the latency of producing each\nelement by iterating over a dataset: \n\n dataset = ...\n dataset = dataset.apply(tf.data.experimental.latency_stats(\"total_bytes\"))\n\nTo associate a `StatsAggregator` with a [`tf.data.Dataset`](../../../../../tf/data/Dataset) object, use\nthe following pattern: \n\n aggregator = tf.data.experimental.StatsAggregator()\n dataset = ...\n\n # Apply `StatsOptions` to associate `dataset` with `aggregator`.\n options = tf.data.Options()\n options.experimental_stats.aggregator = aggregator\n dataset = dataset.with_options(options)\n\nTo get a protocol buffer summary of the currently aggregated statistics,\nuse the `StatsAggregator.get_summary()` tensor. The easiest way to do this\nis to add the returned tensor to the `tf.GraphKeys.SUMMARIES` collection,\nso that the summaries will be included with any existing summaries. \n\n aggregator = tf.data.experimental.StatsAggregator()\n # ...\n stats_summary = aggregator.get_summary()\n tf.compat.v1.add_to_collection(tf.GraphKeys.SUMMARIES, stats_summary)\n\n| **Note:** This interface is experimental and expected to change. In particular, we expect to add other implementations of `StatsAggregator` that provide different ways of exporting statistics, and add more types of statistics.\n\nMethods\n-------\n\n### `get_summary`\n\n[View source](https://github.com/tensorflow/tensorflow/blob/v2.0.0/tensorflow/python/data/experimental/ops/stats_aggregator.py#L130-L140) \n\n get_summary()\n\nReturns a string [`tf.Tensor`](../../../../../tf/Tensor) that summarizes the aggregated statistics.\n\nThe returned tensor will contain a serialized [`tf.compat.v1.summary.Summary`](../../../../../tf/compat/v1/Summary)\nprotocol\nbuffer, which can be used with the standard TensorBoard logging facilities.\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n| Returns ||\n|---|---|\n| A scalar string [`tf.Tensor`](../../../../../tf/Tensor) that summarizes the aggregated statistics. ||\n\n\u003cbr /\u003e"]]
View source on GitHub