Pooling

#include <popnn/Pooling.hpp>

Support for pooling operations.

namespace popnn

Functions used in neural networks.

namespace pooling

Functions for pooling operations.

Functions

std::ostream &operator<<(std::ostream &o, const PoolParams &params)
const char *asString(const PoolingType &method)
std::vector<std::size_t> getOutputFieldShape(const PoolParams &params)
uint64_t getFwdFlops(const PoolParams &params)
uint64_t getBwdFlops(const PoolParams &params)
double getFwdPerfectCycleCount(const poplar::Graph &graph, const PoolParams &params)
double getBwdPerfectCycleCount(const poplar::Graph &graph, const PoolParams &params)
poplar::Tensor pool(poplar::Graph &graph, const PoolParams &params, const poplar::Tensor &in, poplar::program::Sequence &prog, const poplar::DebugContext &debugContext = {}, const poplar::OptionFlags &options = {})

Add a pooling operation to the graph.

This performs a pooling over the spatial dimensions […]. The shape of the input should be [batchSize * numChannels * ...].

Parameters

  • graph – The graph that the operation will be added to.

  • params – The pooling parameters.

  • in – The input tensor.

  • prog – The program sequence to append the operation to.

  • debugContext – Optional debug information.

  • options – Pooling options (not currently used).

Returns

A tensor with the results of the pooling operation.

poplar::Tensor poolInputGradient(poplar::Graph &graph, const PoolParams &params, const poplar::Tensor &in, const poplar::Tensor &pooled, const poplar::Tensor &pooledGradient, bool useScaledGradient, poplar::program::Sequence &prog, const poplar::DebugContext &debugContext = {}, const poplar::OptionFlags &options = {})

Calculate the gradient with respect to the input of a pooling operation given the gradient of the output.

While this function can be used for maximum, average or sum pooling, poolInputGradient(poplar::Graph&, const PoolParams&, const unsigned, const poplar::Tensor&, poplar::program::Sequence&, const poplar::DebugContext&, const poplar::OptionFlags&) is recommended for average or sum pooling.

This performs a pooling over the spatial dimensions […]. The shape of the input should be [batchSize * numChannels * ...].

Parameters

  • graph – The graph the operation will be added to.

  • params – The pooling parameters.

  • in – The forward activations tensor input to pooling.

  • pooled – The output of pooling in the forward pass.

  • pooledGradient – The gradients to the pooling operation.

  • useScaledGradient – Use a scaled gradient if set to true. Otherwise, the gradient is propagated to all the positions which matched the pooled value in the forward pass.

  • prog – The program sequence to append the operation to.

  • debugContext – Optional debug information.

  • options – The pooling options. See pool().

Returns

A tensor with the results of the pooling operation.

poplar::Tensor poolInputGradient(poplar::Graph &graph, const PoolParams &params, const unsigned fwdChansPerGroup, const poplar::Tensor &pooledGradient, poplar::program::Sequence &prog, const poplar::DebugContext &debugContext = {}, const poplar::OptionFlags &options = {})

Calculate the gradient with respect to the input of a pooling operation given the gradient of the output.

This function should be used for average and sum pooling.

This performs a pooling over the spatial dimensions […]. The shape of the output will be [batchSize * numChannels * ...].

Parameters

  • graph – The graph the operation will be added to.

  • params – The pooling parameters.

  • fwdChansPerGroup – Used in creating the output tensor.

  • pooledGradient – The gradients to the pooling operation.

  • prog – The program sequence to append the operation to.

  • debugContext – Optional debug information.

  • options – The pooling options. See pool().

Returns

A tensor with the results of the pooling operation.

struct PoolParams

Public Functions

inline PoolParams(PoolingType poolingType, std::vector<std::size_t> inputFieldShape, std::vector<std::size_t> kernelShape, std::vector<unsigned> stride, std::vector<int> inputTruncationOrPaddingLower, std::vector<int> inputTruncationOrPaddingUpper, std::size_t numChannels, std::size_t batchSize, poplar::Type dType)
inline std::size_t getNumFieldDims() const
std::vector<std::size_t> getOutputFieldShape() const

Public Members

PoolingType poolingType

The type of pooling to be performed (for example, maximum, sum or average).

std::vector<std::size_t> inputFieldShape

The input shape not including the dimensions for batch size batchSize and number of channels numChannels.

std::vector<std::size_t> kernelShape

The shape of the pooling kernel.

std::vector<unsigned> stride

The stride per input dimension.

std::vector<int> inputTruncationOrPaddingLower

The lower padding (for values >0) or truncation (for values <0) for each input dimension specifying the padding/truncation at the start of the dimension.

std::vector<int> inputTruncationOrPaddingUpper

The upper padding (for values >0) or truncation (for values <0) for each input dimension specifying the padding/truncation at the end of the dimension.

std::size_t numChannels

The number of channels in the input tensor.

std::size_t batchSize

The batch size of the input tensor.

poplar::Type dType

The output type.