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)==> 3tf.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.
[null,null,["Last updated 2020-10-01 UTC."],[],[],null,["# tf.compat.v1.sparse_reduce_sum\n\n\u003cbr /\u003e\n\n|-------------------------------------------------------------------------------------------------------------------------------|\n| [View source on GitHub](https://github.com/tensorflow/tensorflow/blob/v2.1.0/tensorflow/python/ops/sparse_ops.py#L1337-L1392) |\n\nComputes the sum of elements across dimensions of a SparseTensor. (deprecated arguments) (deprecated arguments)\n\n#### View aliases\n\n\n**Compat aliases for migration**\n\nSee\n[Migration guide](https://www.tensorflow.org/guide/migrate) for\nmore details.\n\n[`tf.compat.v1.sparse.reduce_sum`](/api_docs/python/tf/compat/v1/sparse_reduce_sum)\n\n\u003cbr /\u003e\n\n tf.compat.v1.sparse_reduce_sum(\n sp_input, axis=None, keepdims=None, reduction_axes=None, keep_dims=None\n )\n\n| **Warning:** SOME ARGUMENTS ARE DEPRECATED: `(keep_dims)`. They will be removed in a future version. Instructions for updating: keep_dims is deprecated, use keepdims instead\n| **Warning:** SOME ARGUMENTS ARE DEPRECATED: `(reduction_axes)`. They will be removed in a future version. Instructions for updating: reduction_axes is deprecated, use axis instead\n\nThis Op takes a SparseTensor and is the sparse counterpart to\n[`tf.reduce_sum()`](../../../tf/math/reduce_sum). In particular, this Op also returns a dense `Tensor`\ninstead of a sparse one.\n\nReduces `sp_input` along the dimensions given in `reduction_axes`. Unless\n`keepdims` is true, the rank of the tensor is reduced by 1 for each entry in\n`reduction_axes`. If `keepdims` is true, the reduced dimensions are retained\nwith length 1.\n\nIf `reduction_axes` has no entries, all dimensions are reduced, and a tensor\nwith a single element is returned. Additionally, the axes can be negative,\nsimilar to the indexing rules in Python.\n\n#### For example:\n\n # 'x' represents [[1, ?, 1]\n # [?, 1, ?]]\n # where ? is implicitly-zero.\n tf.sparse.reduce_sum(x) ==\u003e 3\n tf.sparse.reduce_sum(x, 0) ==\u003e [1, 1, 1]\n tf.sparse.reduce_sum(x, 1) ==\u003e [2, 1] # Can also use -1 as the axis.\n tf.sparse.reduce_sum(x, 1, keepdims=True) ==\u003e [[2], [1]]\n tf.sparse.reduce_sum(x, [0, 1]) ==\u003e 3\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n| Args ---- ||\n|------------------|--------------------------------------------------------------------------------------------|\n| `sp_input` | The SparseTensor to reduce. Should have numeric type. |\n| `axis` | The dimensions to reduce; list or scalar. If `None` (the default), reduces all dimensions. |\n| `keepdims` | If true, retain reduced dimensions with length 1. |\n| `reduction_axes` | Deprecated name of `axis`. |\n| `keep_dims` | Deprecated alias for `keepdims`. |\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n| Returns ------- ||\n|---|---|\n| The reduced Tensor. ||\n\n\u003cbr /\u003e"]]
View source on GitHub