提交 17c697c7 编写于 作者: G gaoyuan

Remove redundancy codes

上级 784e242b
...@@ -110,7 +110,7 @@ sum_to_one_norm ...@@ -110,7 +110,7 @@ sum_to_one_norm
:noindex: :noindex:
cross_channel_norm cross_channel_norm
--------------- ------------------
.. automodule:: paddle.v2.layer .. automodule:: paddle.v2.layer
:members: cross_channel_norm :members: cross_channel_norm
:noindex: :noindex:
......
...@@ -78,7 +78,6 @@ void CrossChannelNormLayer::backward(const UpdateCallback& callback) { ...@@ -78,7 +78,6 @@ void CrossChannelNormLayer::backward(const UpdateCallback& callback) {
Matrix::resizeOrCreate(sampleBuffer_, channels_, spatialDim, false, useGpu_); Matrix::resizeOrCreate(sampleBuffer_, channels_, spatialDim, false, useGpu_);
scaleDiff_->zeroMem(); scaleDiff_->zeroMem();
for (size_t i = 0; i < batchSize; i++) { for (size_t i = 0; i < batchSize; i++) {
spatialBuffer_->zeroMem();
// propagate to param. // propagate to param.
MatrixPtr dataBufferTmp = MatrixPtr dataBufferTmp =
Matrix::create(dataBuffer_->getData() + i * dataDim, Matrix::create(dataBuffer_->getData() + i * dataDim,
......
...@@ -68,7 +68,7 @@ public: ...@@ -68,7 +68,7 @@ public:
/** /**
* This layer applys normalization across the channels of each sample to a * This layer applys normalization across the channels of each sample to a
* conv layer's output, and scales the output by a group of trainable factors * conv layer's output, and scales the output by a group of trainable factors
* whose equal to the number of channels. * whose dimensions equal to the number of channels.
* - Input: One and only one input layer are accepted. * - Input: One and only one input layer are accepted.
* - Output: The normalized data of the input data. * - Output: The normalized data of the input data.
* Reference: * Reference:
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册