This class implements a layer which performs max pooling on a set of three-dimensional multi-channel images.
void SetFilterHeight( int filterHeight );
void SetFilterWidth( int filterWidth );
void SetFilterDepth( int filterDepth );
Sets the filter size.
void SetStrideHeight( int strideHeight );
void SetStrideWidth( int strideWidth );
void SetStrideDepth( int strideDepth );
Sets the filter stride. The default value is 1
.
The layer has no trainable parameters.
The single input accepts a blob of the following dimensions:
BatchLength * BatchWidth * ListSize
- the number of images in the setHeight
- images' heightWidth
- images' widthDepth
- images' depthChannels
- the number of channels the image format uses
The single output contains a blob of the dimensions:
BatchLength
is equal to the inputBatchLength
.BatchWidth
is equal to the inputBatchWidth
.ListSize
is equal to the inputListSize
.Height
can be calculated from the inputHeight
as(Height - FilterHeight)/StrideHeight + 1
.Width
can be calculated from the inputWidth
as(Width - FilterWidth)/StrideWidth + 1
.Depth
can be calculated from the inputDepth
as(Depth - FilterDepth)/StrideDepth + 1
.Channels
is equal to the inputChannels
.