tf.compat.v1.sparse_segment_mean

Computes the mean along sparse segments of a tensor.

tf.compat.v1.sparse_segment_mean(
    data,
    indices,
    segment_ids,
    name=None,
    num_segments=None,
    sparse_gradient=False
)

Read the section on segmentation for an explanation of segments.

Like tf.math.segment_mean, but segment_ids can have rank less than data's first dimension, selecting a subset of dimension 0, specified by indices. segment_ids is allowed to have missing ids, in which case the output will be zeros at those indices. In those cases num_segments is used to determine the size of the output.

data A Tensor with data that will be assembled in the output.
indices A 1-D Tensor with indices into data. Has same rank as segment_ids.
segment_ids A 1-D Tensor with indices into the output Tensor. Values should be sorted and can be repeated.
name A name for the operation (optional).
num_segments An optional int32 scalar. Indicates the size of the output Tensor.
sparse_gradient An optional bool. Defaults to False. If True, the gradient of this function will be sparse (IndexedSlices) instead of dense (Tensor). The sparse gradient will contain one non-zero row for each unique index in indices.

A tensor of the shape as data, except for dimension 0 which has size k, the number of segments specified via num_segments or inferred for the last element in segments_ids.