PLearn 0.1
|
Generate samples from a mixture of two gaussians. More...
#include <DichotomizeDond2DiscreteVariables.h>
Public Member Functions | |
DichotomizeDond2DiscreteVariables () | |
Default constructor. | |
int | outputsize () const |
SUBCLASS WRITING: override this so that it returns the size of this learner's output, as a function of its inputsize(), targetsize() and set options. | |
void | train () |
*** SUBCLASS WRITING: *** | |
void | computeOutput (const Vec &, Vec &) const |
*** SUBCLASS WRITING: *** | |
void | computeCostsFromOutputs (const Vec &, const Vec &, const Vec &, Vec &) const |
*** SUBCLASS WRITING: *** | |
TVec< string > | getTestCostNames () const |
*** SUBCLASS WRITING: *** | |
TVec< string > | getTrainCostNames () const |
*** SUBCLASS WRITING: *** | |
VMat | getOutputFile () |
virtual string | classname () const |
virtual OptionList & | getOptionList () const |
virtual OptionMap & | getOptionMap () const |
virtual RemoteMethodMap & | getRemoteMethodMap () const |
virtual DichotomizeDond2DiscreteVariables * | deepCopy (CopiesMap &copies) const |
virtual void | build () |
Finish building the object; just call inherited::build followed by build_() | |
virtual void | makeDeepCopyFromShallowCopy (CopiesMap &copies) |
Transforms a shallow copy into a deep copy. | |
Static Public Member Functions | |
static string | _classname_ () |
static OptionList & | _getOptionList_ () |
static RemoteMethodMap & | _getRemoteMethodMap_ () |
static Object * | _new_instance_for_typemap_ () |
static bool | _isa_ (const Object *o) |
static void | _static_initialize_ () |
static const PPath & | declaringFile () |
Public Attributes | |
TVec< pair< string, TVec< pair < real, real > > > > | discrete_variable_instructions |
### declare public option fields (such as build options) here Start your comments with Doxygen-compatible comments such as //! | |
string | output_path |
The file path for the fixed output file. | |
Static Public Attributes | |
static StaticInitializer | _static_initializer_ |
Static Protected Member Functions | |
static void | declareOptions (OptionList &ol) |
Declares the class options. | |
Private Types | |
typedef PLearner | inherited |
Private Member Functions | |
void | build_ () |
This does the actual building. | |
void | dichotomizeDiscreteVariables () |
Private Attributes | |
VMat | output_file |
Generate samples from a mixture of two gaussians.
Definition at line 52 of file DichotomizeDond2DiscreteVariables.h.
typedef PLearner PLearn::DichotomizeDond2DiscreteVariables::inherited [private] |
Reimplemented from PLearn::PLearner.
Definition at line 54 of file DichotomizeDond2DiscreteVariables.h.
PLearn::DichotomizeDond2DiscreteVariables::DichotomizeDond2DiscreteVariables | ( | ) |
string PLearn::DichotomizeDond2DiscreteVariables::_classname_ | ( | ) | [static] |
Reimplemented from PLearn::PLearner.
Definition at line 52 of file DichotomizeDond2DiscreteVariables.cc.
OptionList & PLearn::DichotomizeDond2DiscreteVariables::_getOptionList_ | ( | ) | [static] |
Reimplemented from PLearn::PLearner.
Definition at line 52 of file DichotomizeDond2DiscreteVariables.cc.
RemoteMethodMap & PLearn::DichotomizeDond2DiscreteVariables::_getRemoteMethodMap_ | ( | ) | [static] |
Reimplemented from PLearn::PLearner.
Definition at line 52 of file DichotomizeDond2DiscreteVariables.cc.
Reimplemented from PLearn::PLearner.
Definition at line 52 of file DichotomizeDond2DiscreteVariables.cc.
Object * PLearn::DichotomizeDond2DiscreteVariables::_new_instance_for_typemap_ | ( | ) | [static] |
Reimplemented from PLearn::Object.
Definition at line 52 of file DichotomizeDond2DiscreteVariables.cc.
StaticInitializer DichotomizeDond2DiscreteVariables::_static_initializer_ & PLearn::DichotomizeDond2DiscreteVariables::_static_initialize_ | ( | ) | [static] |
Reimplemented from PLearn::PLearner.
Definition at line 52 of file DichotomizeDond2DiscreteVariables.cc.
void PLearn::DichotomizeDond2DiscreteVariables::build | ( | ) | [virtual] |
Finish building the object; just call inherited::build followed by build_()
Reimplemented from PLearn::PLearner.
Definition at line 93 of file DichotomizeDond2DiscreteVariables.cc.
{ // ### Nothing to add here, simply calls build_(). inherited::build(); build_(); }
void PLearn::DichotomizeDond2DiscreteVariables::build_ | ( | ) | [private] |
This does the actual building.
Reimplemented from PLearn::PLearner.
Definition at line 103 of file DichotomizeDond2DiscreteVariables.cc.
References PLearn::endl().
{ MODULE_LOG << "build_() called" << endl; if (train_set) { dichotomizeDiscreteVariables(); } }
string PLearn::DichotomizeDond2DiscreteVariables::classname | ( | ) | const [virtual] |
Reimplemented from PLearn::Object.
Definition at line 52 of file DichotomizeDond2DiscreteVariables.cc.
void PLearn::DichotomizeDond2DiscreteVariables::computeCostsFromOutputs | ( | const Vec & | input, |
const Vec & | output, | ||
const Vec & | target, | ||
Vec & | costs | ||
) | const [virtual] |
*** SUBCLASS WRITING: ***
This should be defined in subclasses to compute the weighted costs from already computed output. The costs should correspond to the cost names returned by getTestCostNames().
NOTE: In exotic cases, the cost may also depend on some info in the input, that's why the method also gets so see it.
Implements PLearn::PLearner.
Definition at line 217 of file DichotomizeDond2DiscreteVariables.cc.
{}
void PLearn::DichotomizeDond2DiscreteVariables::computeOutput | ( | const Vec & | input, |
Vec & | output | ||
) | const [virtual] |
*** SUBCLASS WRITING: ***
This should be defined in subclasses to compute the output from the input.
Reimplemented from PLearn::PLearner.
Definition at line 216 of file DichotomizeDond2DiscreteVariables.cc.
{}
void PLearn::DichotomizeDond2DiscreteVariables::declareOptions | ( | OptionList & | ol | ) | [static, protected] |
Declares the class options.
Reimplemented from PLearn::PLearner.
Definition at line 64 of file DichotomizeDond2DiscreteVariables.cc.
References PLearn::OptionBase::buildoption, PLearn::declareOption(), discrete_variable_instructions, and output_path.
{ declareOption(ol, "discrete_variable_instructions", &DichotomizeDond2DiscreteVariables::discrete_variable_instructions, OptionBase::buildoption, "The instructions to dichotomize the variables in the form of field_name : TVec<pair>.\n" "The pairs are values from : to, each creating a 0, 1 variable.\n" "Variables with no specification will be kept as_is.\n"); declareOption(ol, "output_path", &DichotomizeDond2DiscreteVariables::output_path, OptionBase::buildoption, "The file path for the fixed output file."); inherited::declareOptions(ol); }
static const PPath& PLearn::DichotomizeDond2DiscreteVariables::declaringFile | ( | ) | [inline, static] |
Reimplemented from PLearn::PLearner.
Definition at line 92 of file DichotomizeDond2DiscreteVariables.h.
:
//##### Protected Member Functions ######################################
DichotomizeDond2DiscreteVariables * PLearn::DichotomizeDond2DiscreteVariables::deepCopy | ( | CopiesMap & | copies | ) | const [virtual] |
Reimplemented from PLearn::PLearner.
Definition at line 52 of file DichotomizeDond2DiscreteVariables.cc.
void PLearn::DichotomizeDond2DiscreteVariables::dichotomizeDiscreteVariables | ( | ) | [private] |
Definition at line 112 of file DichotomizeDond2DiscreteVariables.cc.
References PLearn::TVec< T >::fill(), PLearn::is_missing(), MISSING_VALUE, PLERROR, PLearn::TVec< T >::size(), PLearn::tostring(), and PLearn::ProgressBar::update().
{ // initialize primary dataset int main_length = train_set->length(); int main_width = train_set->width(); Vec main_input(main_width); TVec<string> main_names(main_width); TVec<int> main_ins(main_width); main_ins.fill(-1); int ins_width = discrete_variable_instructions.size(); main_names << train_set->fieldNames(); for (int ins_col = 0; ins_col < ins_width; ins_col++) { int main_col; for (main_col = 0; main_col < main_width; main_col++) { if (discrete_variable_instructions[ins_col].first == main_names[main_col]) break; } if (main_col >= main_width) PLERROR("In DichotomizeDond2DiscreteVariables: no field with this name in data set: %s", (discrete_variable_instructions[ins_col].first).c_str()); else main_ins[main_col] = ins_col; } // initialize output datasets int output_length = main_length; int output_width = 0; for (int main_col = 0; main_col < main_width; main_col++) { if (main_ins[main_col] < 0) output_width += 1; else { TVec<pair<real, real> > instruction_ptr = discrete_variable_instructions[main_ins[main_col]].second; output_width += instruction_ptr.size(); } } TVec<string> output_names(output_width); int output_col = 0; for (int main_col = 0; main_col < main_width; main_col++) { if (main_ins[main_col] < 0) { output_names[output_col] = main_names[main_col]; output_col += 1; } else { TVec<pair<real, real> > instruction_ptr = discrete_variable_instructions[main_ins[main_col]].second; if (instruction_ptr.size() == 0) continue; for (int ins_col = 0; ins_col < instruction_ptr.size(); ins_col++) { output_names[output_col] = main_names[main_col] + "_" + tostring(instruction_ptr[ins_col].first) + "_" + tostring(instruction_ptr[ins_col].second); output_col += 1; } } } output_file = new FileVMatrix(output_path + ".pmat", output_length, output_names); output_file->defineSizes(output_width, 0, 0); //Now, we can process the discrete variables. ProgressBar* pb = 0; pb = new ProgressBar( "Dichotomizing the discrete variables", main_length); Vec output_record(output_width); for (int main_row = 0; main_row < main_length; main_row++) { train_set->getRow(main_row, main_input); output_col = 0; for (int main_col = 0; main_col < main_width; main_col++) { if (main_ins[main_col] < 0) { output_record[output_col] = main_input[main_col]; output_col += 1; } else { TVec<pair<real, real> > instruction_ptr = discrete_variable_instructions[main_ins[main_col]].second; if (instruction_ptr.size() == 0) continue; for (int ins_col = 0; ins_col < instruction_ptr.size(); ins_col++) { if (is_missing(main_input[main_col])) output_record[output_col] = MISSING_VALUE; else if (main_input[main_col] < instruction_ptr[ins_col].first || main_input[main_col] > instruction_ptr[ins_col].second) output_record[output_col] = 0.0; else output_record[output_col] = 1.0; output_col += 1; } } } output_file->putRow(main_row, output_record); pb->update( main_row ); } delete pb; }
OptionList & PLearn::DichotomizeDond2DiscreteVariables::getOptionList | ( | ) | const [virtual] |
Reimplemented from PLearn::Object.
Definition at line 52 of file DichotomizeDond2DiscreteVariables.cc.
OptionMap & PLearn::DichotomizeDond2DiscreteVariables::getOptionMap | ( | ) | const [virtual] |
Reimplemented from PLearn::Object.
Definition at line 52 of file DichotomizeDond2DiscreteVariables.cc.
VMat PLearn::DichotomizeDond2DiscreteVariables::getOutputFile | ( | ) |
Definition at line 206 of file DichotomizeDond2DiscreteVariables.cc.
{ return output_file; }
RemoteMethodMap & PLearn::DichotomizeDond2DiscreteVariables::getRemoteMethodMap | ( | ) | const [virtual] |
Reimplemented from PLearn::Object.
Definition at line 52 of file DichotomizeDond2DiscreteVariables.cc.
TVec< string > PLearn::DichotomizeDond2DiscreteVariables::getTestCostNames | ( | ) | const [virtual] |
*** SUBCLASS WRITING: ***
This should return the names of the costs computed by computeCostsFromOutputs.
Implements PLearn::PLearner.
Definition at line 218 of file DichotomizeDond2DiscreteVariables.cc.
References PLearn::TVec< T >::append().
{ TVec<string> result; result.append( "MSE" ); return result; }
TVec< string > PLearn::DichotomizeDond2DiscreteVariables::getTrainCostNames | ( | ) | const [virtual] |
*** SUBCLASS WRITING: ***
This should return the names of the objective costs that the train method computes and for which it updates the VecStatsCollector train_stats.
Implements PLearn::PLearner.
Definition at line 224 of file DichotomizeDond2DiscreteVariables.cc.
References PLearn::TVec< T >::append().
{ TVec<string> result; result.append( "MSE" ); return result; }
void PLearn::DichotomizeDond2DiscreteVariables::makeDeepCopyFromShallowCopy | ( | CopiesMap & | copies | ) | [virtual] |
Transforms a shallow copy into a deep copy.
Reimplemented from PLearn::PLearner.
Definition at line 82 of file DichotomizeDond2DiscreteVariables.cc.
References PLearn::deepCopyField().
{ deepCopyField(discrete_variable_instructions, copies); deepCopyField(output_path, copies); inherited::makeDeepCopyFromShallowCopy(copies); }
int PLearn::DichotomizeDond2DiscreteVariables::outputsize | ( | ) | const [virtual] |
SUBCLASS WRITING: override this so that it returns the size of this learner's output, as a function of its inputsize(), targetsize() and set options.
Implements PLearn::PLearner.
Definition at line 211 of file DichotomizeDond2DiscreteVariables.cc.
{return 0;}
void PLearn::DichotomizeDond2DiscreteVariables::train | ( | ) | [virtual] |
*** SUBCLASS WRITING: ***
The role of the train method is to bring the learner up to stage==nstages, updating the stats with training costs measured on-line in the process.
TYPICAL CODE:
static Vec input; // static so we don't reallocate/deallocate memory each time... static Vec target; // (but be careful that static means shared!) input.resize(inputsize()); // the train_set's inputsize() target.resize(targetsize()); // the train_set's targetsize() real weight; if(!train_stats) // make a default stats collector, in case there's none train_stats = new VecStatsCollector(); if(nstages<stage) // asking to revert to a previous stage! forget(); // reset the learner to stage=0 while(stage<nstages) { // clear statistics of previous epoch train_stats->forget(); //... train for 1 stage, and update train_stats, // using train_set->getSample(input, target, weight); // and train_stats->update(train_costs) ++stage; train_stats->finalize(); // finalize statistics for this epoch }
Implements PLearn::PLearner.
Definition at line 212 of file DichotomizeDond2DiscreteVariables.cc.
References PLERROR.
{ PLERROR("DichotomizeDond2DiscreteVariables: we are done here"); }
Reimplemented from PLearn::PLearner.
Definition at line 92 of file DichotomizeDond2DiscreteVariables.h.
TVec< pair<string, TVec< pair<real, real> > > > PLearn::DichotomizeDond2DiscreteVariables::discrete_variable_instructions |
### declare public option fields (such as build options) here Start your comments with Doxygen-compatible comments such as //!
The instructions to fix the binary variables in the form of field_name : instruction. Supported instructions are 9_is_one, not_0_is_one, not_missing_is_one, not_1000_is_one. Variables with no specification will be kept as_is.
Definition at line 66 of file DichotomizeDond2DiscreteVariables.h.
Referenced by declareOptions().
Definition at line 120 of file DichotomizeDond2DiscreteVariables.h.
The file path for the fixed output file.
Definition at line 69 of file DichotomizeDond2DiscreteVariables.h.
Referenced by declareOptions().