W3cubDocs

/TensorFlow 2.4

tf.sparse.softmax

Applies softmax to a batched N-D SparseTensor.

The inputs represent an N-D SparseTensor with logical shape [..., B, C] (where N >= 2), and with indices sorted in the canonical lexicographic order.

This op is equivalent to applying the normal tf.nn.softmax() to each innermost logical submatrix with shape [B, C], but with the catch that the implicitly zero elements do not participate. Specifically, the algorithm is equivalent to:

(1) Applies tf.nn.softmax() to a densified view of each innermost submatrix with shape [B, C], along the size-C dimension; (2) Masks out the original implicitly-zero locations; (3) Renormalizes the remaining elements.

Hence, the SparseTensor result has exactly the same non-zero indices and shape.

Example:

# First batch:
# [?   e.]
# [1.  ? ]
# Second batch:
# [e   ? ]
# [e   e ]
shape = [2, 2, 2]  # 3-D SparseTensor
values = np.asarray([[[0., np.e], [1., 0.]], [[np.e, 0.], [np.e, np.e]]])
indices = np.vstack(np.where(values)).astype(np.int64).T

result = tf.sparse.softmax(tf.sparse.SparseTensor(indices, values, shape))
# ...returning a 3-D SparseTensor, equivalent to:
# [?   1.]     [1    ?]
# [1.  ? ] and [.5  .5]
# where ? means implicitly zero.
Args
sp_input N-D SparseTensor, where N >= 2.
name optional name of the operation.
Returns
output N-D SparseTensor representing the results.

© 2020 The TensorFlow Authors. All rights reserved.
Licensed under the Creative Commons Attribution License 3.0.
Code samples licensed under the Apache 2.0 License.
https://www.tensorflow.org/versions/r2.4/api_docs/python/tf/sparse/softmax