tf.raw_ops.GetMinibatchSplitsWithPhysicalReplica 

tf.raw_ops.GetMinibatchSplitsWithPhysicalReplica(
    program_key,
    row_ids,
    col_ids,
    gains,
    sample_count,
    num_replica,
    table_vocab_size,
    feature_width,
    num_sc_per_chip,
    table_name,
    mini_batch_splits,
    name=None
)

program_key A Tensor of type string.
row_ids A Tensor of type int32.
col_ids A Tensor of type int32.
gains A Tensor of type float32.
sample_count An int that is >= 1.
num_replica An int that is >= 1.
table_vocab_size An int that is >= 1.
feature_width An int that is >= 1.
num_sc_per_chip An int that is >= 1.
table_name A string.
mini_batch_splits A string.
name A name for the operation (optional).

A tuple of Tensor objects (sorted_row_ids, sorted_col_ids, sorted_gains, splits, id_counts, max_ids, max_uniques).
sorted_row_ids A Tensor of type int32.
sorted_col_ids A Tensor of type int32.
sorted_gains A Tensor of type float32.
splits A Tensor of type int64.
id_counts A Tensor of type int32.
max_ids A Tensor of type int32.
max_uniques A Tensor of type int32.