parent
325512c3a1
commit
17f0dde5ba
|
|
@ -153,9 +153,7 @@ rank as the input. This is the reverse transformation of Space2Batch.
|
|||
Carries out batch normalization as described in the paper
|
||||
https://arxiv.org/abs/1502.03167.
|
||||
|
||||
$$\hat x_i\leftarrow \frac{x_i-\mu_\mathcal{B}}{\sqrt{\sigma_\mathcal{B}^2+\epsilon}}$$
|
||||
|
||||
$$y_i=\gamma\hat x_i+\beta\equiv BN_{\gamma,\beta}(x_i)$$
|
||||
Y = (X - Mean) / Sqrt( Var + Eps) * Gama + Beta
|
||||
|
||||
<a class="mk-toclify" id="clip"></a>
|
||||
## Clip
|
||||
|
|
@ -767,4 +765,4 @@ how to bind input/output: take unidirectional_sequence_lstm_test.cc
|
|||
|
||||
Unpacks the given dimension of a rank-R tensor into rank-(R-1) tensors.
|
||||
- axis : An int. The axis to unstack along. Defaults to the first dimension.
|
||||
Negative values wrap around, so the valid range is [-R, R).
|
||||
Negative values wrap around, so the valid range is [-R, R).
|
||||
|
|
|
|||
Loading…
Reference in New Issue