PLearn 0.1
Public Member Functions | Static Public Member Functions | Public Attributes | Static Public Attributes | Static Protected Member Functions | Protected Attributes | Private Types | Private Member Functions
PLearn::TestInTrainSplitter Class Reference

#include <TestInTrainSplitter.h>

Inheritance diagram for PLearn::TestInTrainSplitter:
Inheritance graph
[legend]
Collaboration diagram for PLearn::TestInTrainSplitter:
Collaboration graph
[legend]

List of all members.

Public Member Functions

 TestInTrainSplitter ()
virtual void build ()
 Post-constructor.
virtual void makeDeepCopyFromShallowCopy (CopiesMap &copies)
 Transforms a shallow copy into a deep copy.
virtual string classname () const
virtual OptionListgetOptionList () const
virtual OptionMapgetOptionMap () const
virtual RemoteMethodMapgetRemoteMethodMap () const
virtual TestInTrainSplitterdeepCopy (CopiesMap &copies) const
virtual void setDataSet (VMat the_dataset)
 Overridden to forward to underlying splitter and reset 'first_source_split'.
virtual int nsplits () const
 Returns the number of available different "splits".
virtual int nSetsPerSplit () const
 Returns the number of sets per split.
virtual void getFirstSplit () const
 Compute the first split of source_splitter and store it in 'first_source_split'.
virtual TVec< VMatgetSplit (int i=0)
 Returns split number i.

Static Public Member Functions

static string _classname_ ()
 Declares name and deepCopy methods.
static OptionList_getOptionList_ ()
static RemoteMethodMap_getRemoteMethodMap_ ()
static Object_new_instance_for_typemap_ ()
static bool _isa_ (const Object *o)
static void _static_initialize_ ()
static const PPathdeclaringFile ()

Public Attributes

real percentage_added
PP< Splittersource_splitter

Static Public Attributes

static StaticInitializer _static_initializer_

Static Protected Member Functions

static void declareOptions (OptionList &ol)
 Declares this class' options.

Protected Attributes

TVec< VMatfirst_source_split
 Store the first split provided by the source_splitter.
int n_left
 The number of samples in the test set that would be left if we didn't make sure they were all added once.
int n_splits_per_source_split
 The number of splits to make for each split given by the source_splitter.
int n_train
 The number of samples in the train and test parts, as given by the source_splitter.
int n_test
int n_to_add
 The number of sample to add each time in the train set.

Private Types

typedef Splitter inherited

Private Member Functions

void build_ ()
 This does the actual building.

Detailed Description

Definition at line 52 of file TestInTrainSplitter.h.


Member Typedef Documentation

Reimplemented from PLearn::Splitter.

Definition at line 57 of file TestInTrainSplitter.h.


Constructor & Destructor Documentation

PLearn::TestInTrainSplitter::TestInTrainSplitter ( )

Definition at line 51 of file TestInTrainSplitter.cc.

    : percentage_added(0.1)
    /* ### Initialize all fields to their default value */
{
    // ...

    // ### You may or may not want to call build_() to finish building the object
    // build_();
}

Member Function Documentation

string PLearn::TestInTrainSplitter::_classname_ ( ) [static]

Declares name and deepCopy methods.

Reimplemented from PLearn::Splitter.

Definition at line 73 of file TestInTrainSplitter.cc.

OptionList & PLearn::TestInTrainSplitter::_getOptionList_ ( ) [static]

Reimplemented from PLearn::Splitter.

Definition at line 73 of file TestInTrainSplitter.cc.

RemoteMethodMap & PLearn::TestInTrainSplitter::_getRemoteMethodMap_ ( ) [static]

Reimplemented from PLearn::Splitter.

Definition at line 73 of file TestInTrainSplitter.cc.

bool PLearn::TestInTrainSplitter::_isa_ ( const Object o) [static]

Reimplemented from PLearn::Splitter.

Definition at line 73 of file TestInTrainSplitter.cc.

Object * PLearn::TestInTrainSplitter::_new_instance_for_typemap_ ( ) [static]

Reimplemented from PLearn::Object.

Definition at line 73 of file TestInTrainSplitter.cc.

StaticInitializer TestInTrainSplitter::_static_initializer_ & PLearn::TestInTrainSplitter::_static_initialize_ ( ) [static]

Reimplemented from PLearn::Splitter.

Definition at line 73 of file TestInTrainSplitter.cc.

void PLearn::TestInTrainSplitter::build ( ) [virtual]

Post-constructor.

The normal implementation should call simply inherited::build(), then this class's build_(). This method should be callable again at later times, after modifying some option fields to change the "architecture" of the object.

Reimplemented from PLearn::Object.

Definition at line 105 of file TestInTrainSplitter.cc.

References PLearn::Object::build(), and build_().

Here is the call graph for this function:

void PLearn::TestInTrainSplitter::build_ ( ) [private]

This does the actual building.

Reimplemented from PLearn::Object.

Definition at line 94 of file TestInTrainSplitter.cc.

Referenced by build().

{
    // ### This method should do the real building of the object,
    // ### according to set 'options', in *any* situation.
    // ### Typical situations include:
    // ###  - Initial building of an object from a few user-specified options
    // ###  - Building of a "reloaded" object: i.e. from the complete set of all serialised options.
    // ###  - Updating or "re-building" of an object after a few "tuning" options have been modified.
    // ### You should assume that the parent class' build_() has already been called.
}

Here is the caller graph for this function:

string PLearn::TestInTrainSplitter::classname ( ) const [virtual]

Reimplemented from PLearn::Object.

Definition at line 73 of file TestInTrainSplitter.cc.

void PLearn::TestInTrainSplitter::declareOptions ( OptionList ol) [static, protected]

Declares this class' options.

Reimplemented from PLearn::Splitter.

Definition at line 75 of file TestInTrainSplitter.cc.

References PLearn::OptionBase::buildoption, PLearn::declareOption(), PLearn::Splitter::declareOptions(), percentage_added, and source_splitter.

{
    // ### Declare all of this object's options here
    // ### For the "flags" of each option, you should typically specify
    // ### one of OptionBase::buildoption, OptionBase::learntoption or
    // ### OptionBase::tuningoption. Another possible flag to be combined with
    // ### is OptionBase::nosave

    declareOption(ol, "percentage_added", &TestInTrainSplitter::percentage_added, OptionBase::buildoption,
                  "The ratio between the number of examples in the test set added to the train set and the\n"
                  "number of examples in the train set.");

    declareOption(ol, "source_splitter", &TestInTrainSplitter::source_splitter, OptionBase::buildoption,
                  "The underlying splitter.");

    // Now call the parent class' declareOptions
    inherited::declareOptions(ol);
}

Here is the call graph for this function:

static const PPath& PLearn::TestInTrainSplitter::declaringFile ( ) [inline, static]

Reimplemented from PLearn::Splitter.

Definition at line 128 of file TestInTrainSplitter.h.

TestInTrainSplitter * PLearn::TestInTrainSplitter::deepCopy ( CopiesMap copies) const [virtual]

Reimplemented from PLearn::Splitter.

Definition at line 73 of file TestInTrainSplitter.cc.

void PLearn::TestInTrainSplitter::getFirstSplit ( ) const [virtual]

Compute the first split of source_splitter and store it in 'first_source_split'.

Definition at line 162 of file TestInTrainSplitter.cc.

References first_source_split, PLearn::TVec< T >::length(), n_test, n_train, and source_splitter.

Referenced by getSplit(), and nsplits().

                                              {
    first_source_split = source_splitter->getSplit(0);
    n_train = first_source_split[0]->length();
    n_test = first_source_split[1]->length();
}

Here is the call graph for this function:

Here is the caller graph for this function:

OptionList & PLearn::TestInTrainSplitter::getOptionList ( ) const [virtual]

Reimplemented from PLearn::Object.

Definition at line 73 of file TestInTrainSplitter.cc.

OptionMap & PLearn::TestInTrainSplitter::getOptionMap ( ) const [virtual]

Reimplemented from PLearn::Object.

Definition at line 73 of file TestInTrainSplitter.cc.

RemoteMethodMap & PLearn::TestInTrainSplitter::getRemoteMethodMap ( ) const [virtual]

Reimplemented from PLearn::Object.

Definition at line 73 of file TestInTrainSplitter.cc.

TVec< VMat > PLearn::TestInTrainSplitter::getSplit ( int  i = 0) [virtual]

Returns split number i.

Implements PLearn::Splitter.

Definition at line 171 of file TestInTrainSplitter.cc.

References first_source_split, getFirstSplit(), i, PLearn::TVec< T >::isEmpty(), PLearn::TVec< T >::length(), PLearn::VMat::length(), n_left, n_splits_per_source_split, n_test, n_to_add, n_train, PLERROR, source_splitter, PLearn::vconcat(), and PLearn::VMat::width().

{
    TVec<VMat> source_split;
    if (first_source_split.isEmpty()) {
        getFirstSplit();
    }
    if (k == 0) {
        source_split = first_source_split;
    } else {
        source_split = source_splitter->getSplit(k / n_splits_per_source_split);
    }
    int n_test_part = k % n_splits_per_source_split;
    int i_test_start = n_test_part * n_to_add;
    VMat train_set = source_split[0];
    VMat test_set = source_split[1];
    if (train_set->length() != n_train || test_set->length() != n_test) {
        PLERROR("In TestInTrainSplitter::getSplit - The train / test sizes have changed!");
    }
    TVec<VMat> result(source_split.length());
    if (n_to_add == 0) {
        // Do not change the split.
        result[0] = train_set;
        result[1] = test_set;
    } else if (n_left == 0 || n_test_part != n_splits_per_source_split - 1) {
        // Easy case: we add the same subset in train that the one for test.
        VMat added_to_train = new SubVMatrix(test_set, i_test_start, 0, n_to_add, test_set->width());
        result[0] = vconcat(train_set, added_to_train);
        result[1] = added_to_train;
    } else {
        // We also take the beginning of the test to fill added_to_train,
        // so that we add the correct number of points in the training set.
        VMat new_test = new SubVMatrix(test_set, i_test_start, 0, n_left, test_set->width());
        result[1] = new_test;
        VMat compl_for_train = new SubVMatrix(test_set, 0, 0, n_to_add - n_left, test_set->width());
        VMat added_to_train = vconcat(new_test, compl_for_train);
        result[0] = vconcat(train_set, added_to_train);
    }
    for (int i = 2; i < result.length(); i++) {
        result[i] = source_split[i];
    }
    return result;
}

Here is the call graph for this function:

void PLearn::TestInTrainSplitter::makeDeepCopyFromShallowCopy ( CopiesMap copies) [virtual]

Transforms a shallow copy into a deep copy.

Reimplemented from PLearn::Splitter.

Definition at line 111 of file TestInTrainSplitter.cc.

References PLearn::Splitter::makeDeepCopyFromShallowCopy(), and PLERROR.

{
    inherited::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("TestInTrainSplitter::makeDeepCopyFromShallowCopy not fully (correctly) implemented yet!");
}

Here is the call graph for this function:

int PLearn::TestInTrainSplitter::nSetsPerSplit ( ) const [virtual]

Returns the number of sets per split.

Implements PLearn::Splitter.

Definition at line 154 of file TestInTrainSplitter.cc.

References source_splitter.

{
    return source_splitter->nSetsPerSplit();
}
int PLearn::TestInTrainSplitter::nsplits ( ) const [virtual]

Returns the number of available different "splits".

Implements PLearn::Splitter.

Definition at line 128 of file TestInTrainSplitter.cc.

References first_source_split, getFirstSplit(), PLearn::TVec< T >::isEmpty(), n_left, n_splits_per_source_split, n_test, n_to_add, n_train, percentage_added, PLERROR, and source_splitter.

{
    if (first_source_split.isEmpty()) {
        getFirstSplit(); // This is just to initialize n_train and n_test.
    }
    n_to_add = int(n_train * percentage_added + 0.5);
    if (n_to_add == 0) {
        // Do NOT add points in the train set.
        n_splits_per_source_split = 1;
        n_left = 0;
    } else {
        n_splits_per_source_split = n_test / n_to_add;
        n_left = n_test % n_to_add;
    }
    if (n_splits_per_source_split == 0) {
        PLERROR("In TestInTrainSplitter::nsplits - Asked to add more test samples than available");
    }
    if (n_left > 0)
        n_splits_per_source_split++;
    int n_total = n_splits_per_source_split * source_splitter->nsplits();
    return n_total;
}

Here is the call graph for this function:

void PLearn::TestInTrainSplitter::setDataSet ( VMat  the_dataset) [virtual]

Overridden to forward to underlying splitter and reset 'first_source_split'.

Reimplemented from PLearn::Splitter.

Definition at line 217 of file TestInTrainSplitter.cc.

References first_source_split, PLearn::TVec< T >::resize(), PLearn::Splitter::setDataSet(), and source_splitter.

                                                     {
    first_source_split.resize(0);
    inherited::setDataSet(the_dataset);
    source_splitter->setDataSet(the_dataset);
}

Here is the call graph for this function:


Member Data Documentation

Reimplemented from PLearn::Splitter.

Definition at line 128 of file TestInTrainSplitter.h.

Store the first split provided by the source_splitter.

This is because we need it the number of points in the train and test parts in the nsplits() method.

Definition at line 72 of file TestInTrainSplitter.h.

Referenced by getFirstSplit(), getSplit(), nsplits(), and setDataSet().

The number of samples in the test set that would be left if we didn't make sure they were all added once.

Definition at line 76 of file TestInTrainSplitter.h.

Referenced by getSplit(), and nsplits().

The number of splits to make for each split given by the source_splitter.

Definition at line 79 of file TestInTrainSplitter.h.

Referenced by getSplit(), and nsplits().

Definition at line 83 of file TestInTrainSplitter.h.

Referenced by getFirstSplit(), getSplit(), and nsplits().

The number of sample to add each time in the train set.

Definition at line 86 of file TestInTrainSplitter.h.

Referenced by getSplit(), and nsplits().

The number of samples in the train and test parts, as given by the source_splitter.

Definition at line 83 of file TestInTrainSplitter.h.

Referenced by getFirstSplit(), getSplit(), and nsplits().

Definition at line 94 of file TestInTrainSplitter.h.

Referenced by declareOptions(), and nsplits().


The documentation for this class was generated from the following files:
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Defines