TrainValidTestSplitter.cc

Go to the documentation of this file.

// -*- C++ -*-

// TrainValidTestSplitter.cc
//
// Copyright (C) 2004 Olivier Delalleau 
// 
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are met:
// 
//  1. Redistributions of source code must retain the above copyright
//     notice, this list of conditions and the following disclaimer.
// 
//  2. Redistributions in binary form must reproduce the above copyright
//     notice, this list of conditions and the following disclaimer in the
//     documentation and/or other materials provided with the distribution.
// 
//  3. The name of the authors may not be used to endorse or promote
//     products derived from this software without specific prior written
//     permission.
// 
// THIS SOFTWARE IS PROVIDED BY THE AUTHORS ``AS IS'' AND ANY EXPRESS OR
// IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
// OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN
// NO EVENT SHALL THE AUTHORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED
// TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
// PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
// LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
// NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
// SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
// 
// This file is part of the PLearn library. For more information on the PLearn
// library, go to the PLearn Web site at www.plearn.org

/* *******************************************************      
   * $Id: TrainValidTestSplitter.cc,v 1.3 2004/07/21 16:30:55 chrish42 Exp $ 
   ******************************************************* */

// Authors: Olivier Delalleau

#include "ConcatRowsVMatrix.h"
#include <plearn/math/random.h>
#include "SelectRowsVMatrix.h"
#include "SubVMatrix.h"
#include "TrainValidTestSplitter.h"

namespace PLearn {
using namespace std;

// TrainValidTestSplitter //
TrainValidTestSplitter::TrainValidTestSplitter() 
  : Splitter(),
    append_train(0),
    append_valid(0),
    n_splits(1),
    n_train(-1),
    n_valid(-1),
    shuffle_valid_and_test(1)
{}

PLEARN_IMPLEMENT_OBJECT(TrainValidTestSplitter,
    "This splitter will basically return [Train+Valid, Test].",
    "The train test returned by the splitter is formed from the first n_train+n_valid\n"
    "samples in the dataset. The other samples are returned in the test set.\n"
    "The validation and test sets (given by the samples after the n_train-th one) can\n"
    "be shuffled in order to get a different validation and test sets at each split.\n"
    "However, the train set (the first n_train samples) remains fixed.");

// declareOptions //
void TrainValidTestSplitter::declareOptions(OptionList& ol)
{
   declareOption(ol, "append_train", &TrainValidTestSplitter::append_train, OptionBase::buildoption,
       "If set to 1, the train set will be appended to each split, after the test set\n"
       "(the train set means the first n_train samples).");

   declareOption(ol, "append_valid", &TrainValidTestSplitter::append_valid, OptionBase::buildoption,
       "If set to 1, the validation set will be appended to each split, after the test set\n"
       "(or the train set if append_train is also set to 1).");

   declareOption(ol, "n_splits", &TrainValidTestSplitter::n_splits, OptionBase::buildoption,
       "The number of splits we want (a value > 1 is useful with shuffle_valid_and_test = 1).");

   declareOption(ol, "n_train", &TrainValidTestSplitter::n_train, OptionBase::buildoption,
       "The number of samples that define the train set, assumed to be at the beginning\n"
       "of the dataset.");

   declareOption(ol, "n_valid", &TrainValidTestSplitter::n_valid, OptionBase::buildoption,
       "The number of samples that define the validation set (they are taken among\n"
       "the samples after the n_train first ones).");
     
   declareOption(ol, "shuffle_valid_and_test", &TrainValidTestSplitter::shuffle_valid_and_test, OptionBase::buildoption,
       "If set to 1, then the part of the dataset after the first n_train ones will\n"
       "be shuffled before taking the validation and test sets. Note that if you want\n"
       "to set it to 0, then using a TrainTestSplitter is probably more appropriate.");
   
  // Now call the parent class' declareOptions
  inherited::declareOptions(ol);
}

// build //
void TrainValidTestSplitter::build()
{
  inherited::build();
  build_();
}

// build_ //
void TrainValidTestSplitter::build_()
{
  if (dataset) {
    if (n_train < 0 || n_valid < 0) {
      PLERROR("In TrainValidTestSplitter::build_ - Please initialize correctly 'n_train' and 'n_valid'");
    }
    int n = dataset->length();
    int n_test = n - n_train - n_valid;
    // Define the train set.
    train_set = new SubVMatrix(dataset, 0, 0, n_train, dataset->width());
    // Precompute all the indices.
    valid_indices.resize(n_splits, n_valid);
    test_indices.resize(n_splits, n_test);
    TVec<int> valid_and_test_indices(n_valid + n_test);
    for (int i = 0; i < n_splits; i++) {
      for (int j = 0; j < n_valid + n_test; j++) {
        valid_and_test_indices[j] = j + n_train;
      }
      if (shuffle_valid_and_test) {
        shuffleElements(valid_and_test_indices);
      }
      valid_indices(i) << valid_and_test_indices.subVec(0, n_valid);
      test_indices(i) << valid_and_test_indices.subVec(n_valid, n_test);
      if (shuffle_valid_and_test) {
        // Now sort the indices for (hopefully) faster access.
        sortElements(valid_indices(i));
        sortElements(test_indices(i));
      }
    }
  }
}

// getSplit //
TVec<VMat> TrainValidTestSplitter::getSplit(int k)
{
  // ### Build and return the kth split .
  TVec<VMat> result(2);
  VMat valid_set = new SelectRowsVMatrix(dataset, valid_indices(k));
  result[0] = vconcat(train_set, valid_set);
  result[1] = new SelectRowsVMatrix(dataset, test_indices(k));
  if (append_train) {
    result.append(train_set);
  }
  if (append_valid) {
    result.append(valid_set);
  }
  return result;
}

// makeDeepCopyFromShallowCopy //
void TrainValidTestSplitter::makeDeepCopyFromShallowCopy(map<const void*, void*>& copies)
{
  Splitter::makeDeepCopyFromShallowCopy(copies);

  // ### Call deepCopyField on all "pointer-like" fields 
  // ### that you wish to be deepCopied rather than 
  // ### shallow-copied.
  // ### ex:
  // deepCopyField(trainvec, copies);

  // ### Remove this line when you have fully implemented this method.
  PLERROR("TrainValidTestSplitter::makeDeepCopyFromShallowCopy not fully (correctly) implemented yet!");
}

// nsplits //
int TrainValidTestSplitter::nsplits() const
{
  // ### Return the number of available splits
  return this->n_splits;
}

// nSetsPerSplit //
int TrainValidTestSplitter::nSetsPerSplit() const
{
  // ### Return the number of sets per split
  int result = 2;
  if (append_train) {
    result++;
  }
  if (append_valid) {
    result++;
  }
  return result;
}

// setDataSet //
void TrainValidTestSplitter::setDataSet(VMat the_dataset) {
  inherited::setDataSet(the_dataset);
  build_(); // To recompute the indices.
}

} // end of namespace PLearn

---