removed unused code

src/mlpack/methods/ann/CMakeLists.txt

@@ -14,7 +14,6 @@ add_subdirectory(layer)
 add_subdirectory(loss_functions)
 add_subdirectory(convolution_rules)
 add_subdirectory(regularizer)
-# add_subdirectory(reduction_rules)
 
 # Add directory name to sources.
 set(DIR_SRCS)

src/mlpack/methods/ann/reduction_rules/CMakeLists.txt

@@ -1,14 +0,0 @@
-# Define the files we need to compile
-# Anything not in this list will not be compiled into mlpack.
-set(SOURCES
-  add_reduction.hpp
-)
-
-# Add directory name to sources.
-set(DIR_SRCS)
-foreach(file ${SOURCES})
-  set(DIR_SRCS ${DIR_SRCS} ${CMAKE_CURRENT_SOURCE_DIR}/${file})
-endforeach()
-# Append sources (with directory name) to list of all mlpack sources (used at
-# the parent scope).
-set(MLPACK_SRCS ${MLPACK_SRCS} ${DIR_SRCS} PARENT_SCOPE)

src/mlpack/methods/ann/reduction_rules/add_reduction.hpp

@@ -1,97 +0,0 @@
-/**
- * @file methods/ann/reduction_rules/add_reduction.hpp
- * @author Shubham Agrawal
- *
- * Reduction of a given network with add reduction rule.
- *
- * mlpack is free software; you may redistribute it and/or modify it under the
- * terms of the 3-clause BSD license.  You should have received a copy of the
- * 3-clause BSD license along with mlpack.  If not, see
- * http://www.opensource.org/licenses/BSD-3-Clause for more information.
- */
-#ifndef MLPACK_METHODS_ANN_REDUCTION_RULES_ADD_REDUCTION_HPP
-#define MLPACK_METHODS_ANN_REDUCTION_RULES_ADD_REDUCTION_HPP
-
-#include <mlpack/prereqs.hpp>
-#include <mlpack/methods/ann/layer/layer.hpp>
-
-namespace mlpack {
-namespace ann /** Artificial Neural Network. */ {
-
-/**
- * This class is used to reduce the network with the add reduction rule.
- * @tparam MatType Type of matrix used.
- */
-template<typename MatType>
-class AddReductionType
-{
- public:
-  /**
-   * Reduce the specified network and store the results in the given
-   * parameter.
-   *
-   * @param layerOutputs Network output that should be reduced.
-   * @param output The network output.
-   */
-  void Reduce(const std::vector<MatType>& layerOutputs,
-              MatType& output)
-  {
-    output.zeros();
-		for (size_t i = 0; i < layerOutputs.size(); i++)
-		{
-			output += layerOutputs[i];
-		}
-  }
-
-	/**
-   * Reduce the specified network and store the results in the given
-   * parameter.
-   *
-   * @param input Network layer wise delta output.
-   * @param networkSize The network size.
-	 * @param layerDeltas The network deltas output.
-   */
-	void UnReduce(
-			const MatType& input,
-			const size_t networkSize,
-			std::vector<MatType>& layerDeltas)
-	{
-		layerDeltas.resize(networkSize, MatType());
-		for (size_t i = 0; i < networkSize; i++)
-		{
-			layerDeltas[i] = input;
-		}
-	}
-
-	std::vector<size_t> ReduceSize(
-			const std::vector<Layer<MatType>*>& network)
-	{
-		if (network.size() == 0)
-		{
-			return std::vector<size_t>();
-		} 
-		else if (network.size() == 1) 
-		{
-			return network[0]->OutputDimensions();
-		}
-		const std::vector<size_t> networkSize = network[0]->OutputDimensions();
-		for (size_t i = 1; i < network.size(); i++)
-		{
-			if (!(networkSize == network[i]->OutputDimensions()))
-			{
-				Log::Fatal << "Network size mismatch. (" << networkSize[0] << ", "
-					<< networkSize[1] << ") != ("
-					<< network[i]->OutputDimensions()[0] << ", " << network[i]->OutputDimensions()[1]
-					<< ")." << std::endl;
-			}
-		}
-		return networkSize;
-	}
-}; // class AddReduction
-
-typedef AddReductionType<arma::mat> AddReduction;
-
-} // namespace ann
-} // namespace mlpack
-
-#endif