Support rectangle input for CNN

paddle/gserver/layers/BatchNormBaseLayer.cpp

@@ -61,15 +61,10 @@ bool BatchNormBaseLayer::init(const LayerMap& layerMap,
 
 void BatchNormBaseLayer::calFeatureMapSize() {
   const ImageConfig& conf = config_.inputs(0).image_conf();
-  if (inputLayers_[0]->getOutput().getFrameHeight() == 0 &&
-      inputLayers_[0]->getOutput().getFrameWidth() == 0) {
-    imgSize_ = conf.img_size();
-    imageH_ = imgSize_;
-    imageW_ = imgSize_;
-  } else {
-    imageH_ = inputLayers_[0]->getOutput().getFrameHeight();
-    imageW_ = inputLayers_[0]->getOutput().getFrameWidth();
-  }
+  imageH_ = inputLayers_[0]->getOutput().getFrameHeight();
+  imageW_ = inputLayers_[0]->getOutput().getFrameWidth();
+  if (imageH_ == 0) imageH_ = conf.img_size_y();
+  if (imageW_ == 0) imageW_ = conf.img_size();
   imgPixels_ = imageH_ * imageW_;
   getOutput().setFrameHeight(imageH_);
   getOutput().setFrameWidth(imageW_);

paddle/gserver/layers/BatchNormBaseLayer.h

@@ -78,9 +78,8 @@ class BatchNormBaseLayer : public Layer {
   MatrixPtr savedMean_;
   MatrixPtr savedInvVar_;
 
-  /// Height or width of input image feature, now height is equal to width.
-  /// imgSize is 1 if the input is fully-connected layer.
-  int imgSize_;
+  /// Height or width of input image feature.
+  /// Both of them are 1 if the input is fully-connected layer.
   int imageH_;
   int imageW_;
   /// Height * Width.

paddle/gserver/layers/BilinearInterpLayer.cpp

@@ -26,15 +26,15 @@ size_t BilinearInterpLayer::getSize() {
 
   const BilinearInterpConfig& conf = config_.inputs(0).bilinear_interp_conf();
   if (inImgH_ == 0) {
-    inImgH_ = conf.img_size_y();
+    inImgH_ = conf.image_conf().img_size_y();
   }
   if (inImgW_ == 0) {
-    inImgW_ = conf.img_size_x();
+    inImgW_ = conf.image_conf().img_size();
   }
 
   outImgH_ = conf.out_size_y();
   outImgW_ = conf.out_size_x();
-  numChannels_ = conf.num_channels();
+  numChannels_ = conf.image_conf().channels();
 
   CHECK(outImgH_ > 0 && outImgW_ > 0);
   CHECK(inImgH_ > 0 && inImgW_ > 0);

paddle/gserver/layers/ConvBaseLayer.cpp

@@ -37,11 +37,13 @@ bool ConvBaseLayer::init(const LayerMap& layerMap,
     filterSizeY_.push_back(conf.filter_size_y());
     filterPixels_.push_back(filterSize_.back() * filterSizeY_.back());
     channels_.push_back(conf.channels());
-    imgSizeH_.push_back(conf.img_size());
+    imgSizeH_.push_back(conf.has_img_size_y() ? conf.img_size_y() :
+                        conf.img_size());
     imgSizeW_.push_back(conf.img_size());
     groups_.push_back(conf.groups());
     filterChannels_.push_back(conf.filter_channels());
-    outputH_.push_back(conf.output_x());
+    outputH_.push_back(conf.has_output_y() ? conf.output_y() :
+                       conf.output_x());
     outputW_.push_back(conf.output_x());
   }
 
@@ -90,11 +92,12 @@ size_t ConvBaseLayer::calOutputSize() {
     for (size_t i = 0; i < inputLayers_.size(); i++) {
        inH.push_back(inputLayers_[i]->getOutput().getFrameHeight());
        inW.push_back(inputLayers_[i]->getOutput().getFrameWidth());
+       const ConvConfig& conf = config_.inputs(i).conv_conf();
        if (isDeconv_) {
          if (inH[i] == 0)
-           inH[i] = config_.inputs(i).conv_conf().output_x();
+           inH[i] = conf.has_output_y() ? conf.output_y() : conf.output_x();
          if (inW[i] == 0)
-           inW[i] = config_.inputs(i).conv_conf().output_x();
+           inW[i] = conf.output_x();
          outH.push_back(
              imageSize(inH[i], filterSizeY_[i], paddingY_[i], strideY_[i],
                        caffeMode_));
@@ -103,9 +106,9 @@ size_t ConvBaseLayer::calOutputSize() {
                        caffeMode_));
        } else {
          if (inH[i] == 0)
-           inH[i] = config_.inputs(i).conv_conf().img_size();
+           inH[i] = conf.has_img_size_y() ? conf.img_size_y() : conf.img_size();
          if (inW[i] == 0)
-           inW[i] = config_.inputs(i).conv_conf().img_size();
+           inW[i] = conf.img_size();
          outH.push_back(
              outputSize(inH[i], filterSizeY_[i], paddingY_[i], strideY_[i],
                         caffeMode_));

paddle/gserver/layers/ConvOperator.cpp

@@ -93,9 +93,9 @@ class ConvOperator : public Operator {
   bool caffeMode_;
   int inputOffset_, outputOffset_, weightOffset_;
   int numFilters_;
-  int padding_, stride_, filterSize_, channels_, imgSize_;
+  int padding_, stride_, filterSize_, channels_, imgSize_, imgSizeY_;
   int paddingY_, strideY_, filterSizeY_;
-  int imgPixels_, filterPixels_, filterChannels_, outputX_, outputs_;
+  int imgPixels_, filterPixels_, filterChannels_, outputX_, outputY_, outputs_;
 
   /// Following member variables are same with CudnnConvLayer.
   /// There is no explanation here.
@@ -144,7 +144,7 @@ void ConvOperator::allocConvWorkSpace(size_t maxWorkSpace) {
 void ConvOperator::reshape(int batchSize) {
   imageH_ = ins_[0]->getFrameHeight();
   imageW_ = ins_[0]->getFrameWidth();
-  if (imageH_ == 0) imageH_ = imgSize_;
+  if (imageH_ == 0) imageH_ = imgSizeY_;
   if (imageW_ == 0) imageW_ = imgSize_;
   outputH_ = outputSize(imageH_, filterSizeY_, paddingY_, strideY_, caffeMode_);
   outputW_ = outputSize(imageW_, filterSize_, padding_, stride_, caffeMode_);
@@ -176,7 +176,10 @@ void ConvOperator::computeConvSizes() {
   hl_create_tensor_descriptor(&inputDesc_);
   int outputX =
       outputSize(imgSize_, filterSize_, padding_, stride_, caffeMode_);
+  int outputY =
+      outputSize(imgSizeY_, filterSizeY_, paddingY_, strideY_, caffeMode_);
   CHECK_EQ(outputX, outputX_);
+  CHECK_EQ(outputY, outputY_);
   hl_create_tensor_descriptor(&outputDesc_);
   hl_create_convolution_descriptor(&convDesc_, inputDesc_, filterDesc_,
                                    paddingY_, padding_, strideY_, stride_);
@@ -208,10 +211,12 @@ void ConvOperator::getConvParams() {
   filterPixels_ = filterSize_ * filterSizeY_;
   channels_ = conf.channels();
   imgSize_ = conf.img_size();
-  imgPixels_ = imgSize_ * imgSize_;
+  imgSizeY_ = conf.has_img_size_y() ? conf.img_size_y() : conf.img_size();
+  imgPixels_ = imgSize_ * imgSizeY_;
   CHECK_EQ(conf.groups(), 1U);
   filterChannels_ = conf.filter_channels();
   outputX_ = conf.output_x();
+  outputY_ = conf.has_output_y() ? conf.output_y() : conf.output_x();
   outputs_ = outputX_ * outputX_;
 }
 

paddle/gserver/layers/ConvProjection.cpp

@@ -47,7 +47,7 @@ void ConvProjection::getConvParams() {
   filterH_ = conf.filter_size_y();
   filterW_ = conf.filter_size();
 
-  configImgH_ = conf.img_size();
+  configImgH_ = conf.has_img_size_y() ? conf.img_size_y() : conf.img_size();
   configImgW_ = conf.img_size();
 
   channels_ = conf.channels();

paddle/gserver/layers/DataLayer.cpp

@@ -48,8 +48,8 @@ void DataLayer::copyDataToOutput(Argument& output) {
       output.ids->copyFrom(*data_.ids);
     }
   }
-  output.setFrameHeight(data_.getFrameHeight());
-  output.setFrameWidth(data_.getFrameWidth());
+  output.setFrameHeight(config_.height());
+  output.setFrameWidth(config_.width());
   output.cpuSequenceDims = data_.cpuSequenceDims;
   output.sequenceStartPositions = data_.sequenceStartPositions;
   output.subSequenceStartPositions = data_.subSequenceStartPositions;

paddle/gserver/layers/ExpandConvBaseLayer.cpp

@@ -30,17 +30,19 @@ bool ExpandConvBaseLayer::init(const LayerMap &layerMap,
    * meaning as in conv, we need to swap channels_ and numFilters here for
    * convTrans, and in other functions too.
    * */
-  int channel;
-  int numFilters;
+
   /* Initialize the projection */
   for (auto &inputConfig : config_.inputs()) {
     const ConvConfig &conf = inputConfig.conv_conf();
-    numFilters = isDeconv_ ? conf.channels() : numFilters_;
+    int numFilters = isDeconv_ ? conf.channels() : numFilters_;
     subM_.push_back(numFilters / conf.groups());
-    subN_.push_back(conf.output_x() * conf.output_x());
-    channel = isDeconv_ ? numFilters_ : conf.channels();
-    subK_.push_back(channel * conf.filter_size() * conf.filter_size() /
-                    conf.groups());
+    subN_.push_back(conf.output_x() *
+                    (conf.has_output_y() ? conf.output_y() : conf.output_x()));
+    int channel = isDeconv_ ? numFilters_ : conf.channels();
+    subK_.push_back(
+        channel * conf.filter_size() *
+        (conf.has_filter_size_y() ? conf.filter_size_y() : conf.filter_size()) /
+        conf.groups());
     /* Consistent caffe mode for multiple input */
     caffeMode_ = conf.caffe_mode();
   }
@@ -107,9 +109,9 @@ void ExpandConvBaseLayer::expandOneFrame(MatrixPtr image, size_t startIdx,
       imgData, 1, imgSizeH_[inIdx] * imgSizeW_[inIdx] * channel, false,
       useGpu_);
   expandInput_->convExpand(*imageTmp, imgSizeH_[inIdx], imgSizeW_[inIdx],
-                           channel, filterSize_[inIdx],
-                           filterSize_[inIdx], stride_[inIdx], stride_[inIdx],
-                           padding_[inIdx], padding_[inIdx],
+                           channel, filterSizeY_[inIdx],
+                           filterSize_[inIdx], strideY_[inIdx], stride_[inIdx],
+                           paddingY_[inIdx], padding_[inIdx],
                            outputH_[inIdx], outputW_[inIdx]);
   imageTmp->clear();
 }
@@ -188,10 +190,10 @@ void ExpandConvBaseLayer::bpropActs(MatrixPtr out, MatrixPtr image,
         imgSizeH_[inpIdx] * imgSizeW_[inpIdx] * channel, false,
         useGpu_);
     vTmp->convShrink(*oneGradTmp, imgSizeH_[inpIdx], imgSizeW_[inpIdx],
-                     channel, filterSize_[inpIdx],
-                     filterSize_[inpIdx], stride_[inpIdx], stride_[inpIdx],
-                     padding_[inpIdx], padding_[inpIdx],
-                     outputH_[inpIdx], outputW_[inpIdx], 1.0f, 1.0f);
+                     channel, filterSizeY_[inpIdx],
+                     filterSize_[inpIdx], strideY_[inpIdx], stride_[inpIdx],
+                     paddingY_[inpIdx], padding_[inpIdx], outputH_[inpIdx],
+                     outputW_[inpIdx], 1.0f, 1.0f);
     vTmp->clear();
     oneGradTmp->clear();
 

paddle/gserver/layers/MaxOutLayer.cpp

@@ -25,10 +25,10 @@ size_t MaxOutLayer::getSize() {
   imgSizeH_ = inputLayers_[0]->getOutput().getFrameHeight();
   imgSizeW_ = inputLayers_[0]->getOutput().getFrameWidth();
   if (imgSizeH_ == 0) {
-    imgSizeH_ = maxoutConf.img_size_y();
+    imgSizeH_ = maxoutConf.image_conf().img_size_y();
   }
   if (imgSizeW_ == 0) {
-    imgSizeW_ = maxoutConf.img_size_x();
+    imgSizeW_ = maxoutConf.image_conf().img_size();
   }
 
   featLen_ = imgSizeH_ * imgSizeW_;
@@ -50,7 +50,7 @@ bool MaxOutLayer::init(const LayerMap& layerMap,
 
   const MaxOutConfig& conf = config_.inputs(0).maxout_conf();
   groups_ = conf.groups();
-  channels_ = conf.channels();
+  channels_ = conf.image_conf().channels();
   CHECK_EQ(channels_ % groups_, 0UL);
   outputChannels_ = channels_ / groups_;
 

paddle/gserver/layers/NormLayer.cpp

@@ -49,6 +49,9 @@ bool ResponseNormLayer::init(const LayerMap& layerMap,
   outputX_ = conf.output_x();
   imgSize_ = conf.img_size();
   denoms_ = NULL;
+
+  outputY_ = conf.has_output_y() ? conf.output_y() : conf.output_x();
+  imgSizeY_ = conf.has_img_size_y() ? conf.img_size_y() : conf.img_size();
   return true;
 }
 

paddle/gserver/layers/NormLayer.h

@@ -50,7 +50,7 @@ class NormLayer : public Layer {
  */
 class ResponseNormLayer : public NormLayer {
 protected:
-  size_t channels_, size_, outputX_, imgSize_;
+  size_t channels_, size_, outputX_, imgSize_, outputY_, imgSizeY_;
   float scale_, pow_;
   MatrixPtr denoms_;
 

paddle/gserver/layers/NormProjectionLayer.cpp

@@ -24,7 +24,7 @@ size_t CMRProjectionNormLayer::getSize() {
   imgSizeH_ = inputLayers_[0]->getOutput().getFrameHeight();
   imgSizeW_ = inputLayers_[0]->getOutput().getFrameWidth();
   if (imgSizeH_ == 0) {
-    imgSizeH_ = imgSize_;
+    imgSizeH_ = imgSizeY_;
   }
   if (imgSizeW_ == 0) {
     imgSizeW_ = imgSize_;

paddle/gserver/layers/SpatialPyramidPoolLayer.cpp

@@ -56,14 +56,14 @@ ProjectionConfig SpatialPyramidPoolLayer::getConfig(size_t imgSizeW,
 size_t SpatialPyramidPoolLayer::getSize() {
   CHECK_EQ(inputLayers_.size(), 1UL);
   size_t layerSize = 0;
-  const SppConfig& sppConf = config_.inputs(0).spp_conf();
+  const ImageConfig& conf = config_.inputs(0).spp_conf().image_conf();
   imgSizeH_ = inputLayers_[0]->getOutput().getFrameHeight();
   imgSizeW_ = inputLayers_[0]->getOutput().getFrameWidth();
   if (imgSizeH_ == 0) {
-    imgSizeH_ = sppConf.has_img_size_y() ? sppConf.img_size_y() : imgSizeW_;
+    imgSizeH_ = conf.has_img_size_y() ? conf.img_size_y() : conf.img_size();
   }
   if (imgSizeW_ == 0) {
-    imgSizeW_ = sppConf.img_size();
+    imgSizeW_ = conf.img_size();
   }
 
   size_t outputH = 1;
@@ -82,9 +82,10 @@ bool SpatialPyramidPoolLayer::init(const LayerMap& layerMap,
   pyramidHeight_ = sppConf.pyramid_height();
   poolType_ = sppConf.pool_type();
 
-  channels_ = sppConf.channels();
-  imgSizeW_ = sppConf.img_size();
-  imgSizeH_ = sppConf.has_img_size_y() ? sppConf.img_size_y() : imgSizeW_;
+  const ImageConfig& imageConf = sppConf.image_conf();
+  channels_ = imageConf.channels();
+  imgSizeW_ = imageConf.img_size();
+  imgSizeH_ = imageConf.has_img_size_y() ? imageConf.img_size_y() : imgSizeW_;
   poolProjections_.reserve(pyramidHeight_);
   projCol_.reserve(pyramidHeight_);
   projOutput_.resize(pyramidHeight_);

paddle/gserver/tests/img_pool_a.conf

@@ -28,7 +28,6 @@ maxpool = img_pool_layer(input=conv,
                          stride_y=2,
                          padding=1,
                          padding_y=2,
-                         img_width=16,
                          pool_type=MaxPooling(),
 )
 avgpool = img_pool_layer(input=conv,
@@ -39,7 +38,6 @@ avgpool = img_pool_layer(input=conv,
                          stride_y=2,
                          padding=1,
                          padding_y=2,
-                         img_width=16,
                          pool_type=AvgPooling(),
 )