/TensorFlow 2.4


Computes the max of elements across dimensions of a SparseTensor.

This Op takes a SparseTensor and is the sparse counterpart to tf.reduce_max(). In particular, this Op also returns a dense Tensor if output_is_sparse is False, or a SparseTensor if output_is_sparse is True.

Note: A gradient is not defined for this function, so it can't be used in training models that need gradient descent.

Reduces sp_input along the dimensions given in axis. Unless keepdims is true, the rank of the tensor is reduced by 1 for each entry in axis. If keepdims is true, the reduced dimensions are retained with length 1.

If axis has no entries, all dimensions are reduced, and a tensor with a single element is returned. Additionally, the axes can be negative, similar to the indexing rules in Python.

The values not defined in sp_input don't participate in the reduce max, as opposed to be implicitly assumed 0 -- hence it can return negative values for sparse axis. But, in case there are no values in axis, it will reduce to 0. See second example below.

For example:

# 'x' represents [[1, ?, 2]
#                 [?, 3, ?]]
# where ? is implicitly-zero.
tf.sparse.reduce_max(x) ==> 3
tf.sparse.reduce_max(x, 0) ==> [1, 3, 2]
tf.sparse.reduce_max(x, 1) ==> [2, 3]  # Can also use -1 as the axis.
tf.sparse.reduce_max(x, 1, keepdims=True) ==> [[2], [3]]
tf.sparse.reduce_max(x, [0, 1]) ==> 3

# 'y' represents [[-7, ?]
#                 [ 4, 3]
#                 [ ?, ?]
tf.sparse.reduce_max(x, 1) ==> [-7, 4, 0]
sp_input The SparseTensor to reduce. Should have numeric type.
axis The dimensions to reduce; list or scalar. If None (the default), reduces all dimensions.
keepdims If true, retain reduced dimensions with length 1.
output_is_sparse If true, returns a SparseTensor instead of a dense Tensor (the default).
name A name for the operation (optional).
The reduced Tensor or the reduced SparseTensor if output_is_sparse is True.

© 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.