experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Fixed documentation for batch_normalization ()

Browse Source
This commit is contained in:
Andre Simpelo 2016-09-30 13:29:51 -07:00 committed by Jonathan Hseu
parent d246219976
commit 89d1d5c449
1 changed files with 4 additions and 4 deletions

8
tensorflow/python/ops/nn.py
View File

@ -936,9 +936,9 @@ def batch_normalization(x,
As described in http://arxiv.org/abs/1502.03167. As described in http://arxiv.org/abs/1502.03167.
Normalizes a tensor by `mean` and `variance`, and applies (optionally) a Normalizes a tensor by `mean` and `variance`, and applies (optionally) a
`scale` \\\\(\gamma\\\\) to it, as well as an `offset` \\\\(\\beta\\\\): `scale` \\(\gamma\\) to it, as well as an `offset` \\(\beta\\):
\\\\(\\frac{\gamma(x-\mu)}{\sigma}+\\beta\\\\) \\(\frac{\gamma(x-\mu)}{\sigma}+\beta\\)
`mean`, `variance`, `offset` and `scale` are all expected to be of one of two `mean`, `variance`, `offset` and `scale` are all expected to be of one of two
shapes: shapes:
@ -964,9 +964,9 @@ def batch_normalization(x,
x: Input `Tensor` of arbitrary dimensionality. x: Input `Tensor` of arbitrary dimensionality.
mean: A mean `Tensor`. mean: A mean `Tensor`.
variance: A variance `Tensor`. variance: A variance `Tensor`.
offset: An offset `Tensor`, often denoted \\\\(\\beta\\\\) in equations, or offset: An offset `Tensor`, often denoted \\(\beta\\) in equations, or
None. If present, will be added to the normalized tensor. None. If present, will be added to the normalized tensor.
scale: A scale `Tensor`, often denoted \\\\(\gamma\\\\) in equations, or scale: A scale `Tensor`, often denoted \\(\gamma\\) in equations, or
`None`. If present, the scale is applied to the normalized tensor. `None`. If present, the scale is applied to the normalized tensor.
variance_epsilon: A small float number to avoid dividing by 0. variance_epsilon: A small float number to avoid dividing by 0.
name: A name for this operation (optional). name: A name for this operation (optional).