This Op takes a SparseTensor and is the sparse counterpart to tf.reduce_sum(). In particular, this Op also returns a dense Tensor instead of a sparse one.

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

If reduction_axes 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.

For example:

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

Args

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.