diff options
| author |  A. Unique TensorFlower <gardener@tensorflow.org> | 2016-12-16 15:56:41 -0800 |
|---|---|---|
| committer | TensorFlower Gardener <gardener@tensorflow.org> | 2016-12-16 16:05:46 -0800 |
| commit | 924097dfbb21a37f75bfe452a69e95a26904f29d (patch) | |
| tree | d8e00d506add781ac6b3e21964f16053f61c8932 | |
| parent | c6e38f57c109d935c625c8452b2d4499fccd2845 (diff) | |
Clarify what matmul does in docs.
The new documentation makes it sound like matmul does element-wise
multiplication. This change makes it clear that matrix multiplication will be
performed.
Change: 142307683
| -rw-r--r-- | tensorflow/python/ops/math_ops.py | 5 |
1 files changed, 4 insertions, 1 deletions
diff --git a/tensorflow/python/ops/math_ops.py b/tensorflow/python/ops/math_ops.py index f6f2044412..0f7727ae68 100644 --- a/tensorflow/python/ops/math_ops.py +++ b/tensorflow/python/ops/math_ops.py @@ -1736,7 +1736,10 @@ def matmul(a, the product of the corresponding matrices in `a` and `b, e.g. if all transpose or adjoint attributes are `False`: - output[..., :, :] = a[..., :, :] * b[..., :, :] , + `output`[..., i, j] = sum_k (`a`[..., i, k] * `b`[..., k, j]), + for all indices i, j. + + Note: This is matrix product, not element-wise product. Raises: |