PLearn 0.1
|
Generate samples from a mixture of two gaussians. More...
#include <MergeDond2Files.h>
Public Member Functions | |
MergeDond2Files () | |
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: *** | |
virtual string | classname () const |
virtual OptionList & | getOptionList () const |
virtual OptionMap & | getOptionMap () const |
virtual RemoteMethodMap & | getRemoteMethodMap () const |
virtual MergeDond2Files * | 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 | |
VMat | external_dataset |
### declare public option fields (such as build options) here Start your comments with Doxygen-compatible comments such as //! | |
TVec< pair< string, string > > | missing_instructions |
The variable missing regeneration instructions in the form of pairs field : instruction. | |
TVec< pair< string, string > > | merge_instructions |
The variable merge instructions in the form of pairs field : instruction. | |
string | merge_path |
The file name of the merge file to be created. | |
int | sec_key |
The column of the merge key in the secondary dataset. | |
int | main_key |
The column of the merge key in the main dataset. | |
int | train_ind |
The column of the indicator of a train record in the main dataset. | |
int | test_ind |
The column of the indicator of a test record in the main dataset. | |
VMat | train_file |
The train file created. | |
VMat | test_file |
The test file created. | |
VMat | unknown_file |
The unknown target file created. | |
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 | mergeFiles () |
void | accumulateVec () |
void | combineAndPut () |
Private Attributes | |
int | sec_length |
int | sec_width |
int | sec_row |
Vec | sec_input |
TVec< string > | sec_names |
TVec< string > | sec_ins |
int | extension_width |
int | ext_col |
TVec< int > | extension_pos |
TVec< string > | extension_names |
Mat | sec_values |
Mat | sec_value_cnt |
TVec< int > | sec_value_ind |
bool | sec_value_found |
real | sec_value_count_max |
int | main_length |
int | main_width |
int | main_row |
Vec | main_input |
TVec< string > | main_names |
TVec< string > | main_ins |
int | primary_width |
TVec< string > | primary_names |
int | train_length |
int | test_length |
int | unknown_length |
int | merge_width |
int | train_row |
int | test_row |
int | unknown_row |
int | merge_col |
Vec | merge_output |
TVec< string > | merge_names |
Generate samples from a mixture of two gaussians.
Definition at line 52 of file MergeDond2Files.h.
typedef PLearner PLearn::MergeDond2Files::inherited [private] |
Reimplemented from PLearn::PLearner.
Definition at line 54 of file MergeDond2Files.h.
PLearn::MergeDond2Files::MergeDond2Files | ( | ) |
string PLearn::MergeDond2Files::_classname_ | ( | ) | [static] |
Reimplemented from PLearn::PLearner.
Definition at line 53 of file MergeDond2Files.cc.
OptionList & PLearn::MergeDond2Files::_getOptionList_ | ( | ) | [static] |
Reimplemented from PLearn::PLearner.
Definition at line 53 of file MergeDond2Files.cc.
RemoteMethodMap & PLearn::MergeDond2Files::_getRemoteMethodMap_ | ( | ) | [static] |
Reimplemented from PLearn::PLearner.
Definition at line 53 of file MergeDond2Files.cc.
Reimplemented from PLearn::PLearner.
Definition at line 53 of file MergeDond2Files.cc.
Object * PLearn::MergeDond2Files::_new_instance_for_typemap_ | ( | ) | [static] |
Reimplemented from PLearn::Object.
Definition at line 53 of file MergeDond2Files.cc.
StaticInitializer MergeDond2Files::_static_initializer_ & PLearn::MergeDond2Files::_static_initialize_ | ( | ) | [static] |
Reimplemented from PLearn::PLearner.
Definition at line 53 of file MergeDond2Files.cc.
void PLearn::MergeDond2Files::accumulateVec | ( | ) | [private] |
Definition at line 322 of file MergeDond2Files.cc.
References PLearn::endl(), PLearn::is_missing(), and PLERROR.
{ for (int sec_col = 0; sec_col < sec_width; sec_col++) { if (is_missing(sec_input[sec_col])) continue; else if (sec_ins[sec_col] == "skip") continue; int ext_col = extension_pos[sec_col]; if (sec_ins[sec_col] == "mean") { sec_values(ext_col, 0) += sec_input[sec_col]; sec_value_cnt(ext_col, 0) += 1.0; } else if (sec_ins[sec_col] == "mode") { sec_value_found = false; for (int sec_value_col = 0; sec_value_col < sec_value_ind[sec_col]; sec_value_col++) { if (sec_values(ext_col, sec_value_col) == sec_input[sec_col]) { sec_value_found = true; sec_value_cnt(ext_col, sec_value_col) += 1; break; } } if (!sec_value_found) { if (sec_value_ind[sec_col] >= 10) { cout << "MergeDond2Files: main file row: " << main_row << " external file row: " << sec_row << endl; PLERROR("MergeDond2Files: we have exceeded the capacity of the value MAT."); } sec_values(ext_col, sec_value_ind[sec_col]) = sec_input[sec_col]; sec_value_ind[sec_col] += 1; } } else if (sec_ins[sec_col] == "present") { sec_value_cnt(ext_col, 0) = 1.0; } } }
void PLearn::MergeDond2Files::build | ( | ) | [virtual] |
Finish building the object; just call inherited::build followed by build_()
Reimplemented from PLearn::PLearner.
Definition at line 142 of file MergeDond2Files.cc.
{ // ### Nothing to add here, simply calls build_(). inherited::build(); build_(); }
void PLearn::MergeDond2Files::build_ | ( | ) | [private] |
This does the actual building.
Reimplemented from PLearn::PLearner.
Definition at line 152 of file MergeDond2Files.cc.
References PLearn::endl(), and PLERROR.
{ MODULE_LOG << "build_() called" << endl; if (train_set) { mergeFiles(); PLERROR("MergeDond2Files: we are done here"); } }
string PLearn::MergeDond2Files::classname | ( | ) | const [virtual] |
Reimplemented from PLearn::Object.
Definition at line 53 of file MergeDond2Files.cc.
void PLearn::MergeDond2Files::combineAndPut | ( | ) | [private] |
Definition at line 364 of file MergeDond2Files.cc.
References PLearn::is_missing(), and MISSING_VALUE.
{ merge_col = 0; for (int main_col = 0; main_col < main_width; main_col++) { if (main_ins[main_col] == "skip") continue; else if (main_ins[main_col] == "as_is") { merge_output[merge_col] = main_input[main_col]; } else if (main_ins[main_col] == "zero_is_missing") { if (main_input[main_col] == 0.0) merge_output[merge_col] = MISSING_VALUE; else merge_output[merge_col] = main_input[main_col]; } else if (main_ins[main_col] == "2436935_is_missing") { if (main_input[main_col] == 2436935.0) merge_output[merge_col] = MISSING_VALUE; else merge_output[merge_col] = main_input[main_col]; } else if (main_ins[main_col] == "present") { if (is_missing(main_input[main_col])) merge_output[merge_col] = 0.0; else merge_output[merge_col] = 1.0; } merge_col +=1; } for (int sec_col = 0; sec_col < sec_width; sec_col++) { if (sec_ins[sec_col] == "skip") continue; int ext_col = extension_pos[sec_col]; if (sec_ins[sec_col] == "mean") { if (sec_value_cnt(ext_col, 0) <= 0.0) merge_output[merge_col] = MISSING_VALUE; else merge_output[merge_col] = sec_values(ext_col, 0) / sec_value_cnt(ext_col, 0); } else if (sec_ins[sec_col] == "mode") { if (sec_value_ind[sec_col] <= 0.0) { merge_output[merge_col] = MISSING_VALUE; continue; } merge_output[merge_col] = sec_values(ext_col, 0); sec_value_count_max = sec_value_cnt(ext_col, 0); for (int sec_value_col = 1; sec_value_col < sec_value_ind[sec_col]; sec_value_col++) { if (sec_value_cnt(ext_col, sec_value_col) >= sec_value_count_max) { merge_output[merge_col] = sec_values(ext_col, sec_value_col); sec_value_count_max = sec_value_cnt(ext_col, sec_value_col); } } } else if (sec_ins[sec_col] == "present") { merge_output[merge_col] = sec_value_cnt(ext_col, 0); } merge_col +=1; } if (main_input[train_ind] > 0.0) { train_file->putRow(train_row, merge_output); train_row += 1; } else if (main_input[test_ind] > 0.0) { test_file->putRow(test_row, merge_output); test_row += 1; } else { unknown_file->putRow(unknown_row, merge_output); unknown_row += 1; } sec_values.clear(); sec_value_cnt.clear(); sec_value_ind.clear(); }
void PLearn::MergeDond2Files::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 447 of file MergeDond2Files.cc.
{}
*** SUBCLASS WRITING: ***
This should be defined in subclasses to compute the output from the input.
Reimplemented from PLearn::PLearner.
Definition at line 446 of file MergeDond2Files.cc.
{}
void PLearn::MergeDond2Files::declareOptions | ( | OptionList & | ol | ) | [static, protected] |
Declares the class options.
Reimplemented from PLearn::PLearner.
Definition at line 65 of file MergeDond2Files.cc.
References PLearn::OptionBase::buildoption, PLearn::declareOption(), external_dataset, PLearn::OptionBase::learntoption, main_key, merge_instructions, merge_path, missing_instructions, sec_key, test_file, test_ind, train_file, train_ind, and unknown_file.
{ declareOption(ol, "external_dataset", &MergeDond2Files::external_dataset, OptionBase::buildoption, "The secondary dataset to merge with the main one.\n" "The main one is provided as the train_set."); declareOption(ol, "missing_instructions", &MergeDond2Files::missing_instructions, OptionBase::buildoption, "The variable missing regeneration instructions in the form of pairs field : instruction.\n" "Supported instructions are skip, as_is, zero_is_missing, 2436935_is_missing, present."); declareOption(ol, "merge_instructions", &MergeDond2Files::merge_instructions, OptionBase::buildoption, "The variable merge instructions in the form of pairs field : instruction.\n" "Supported instructions are skip, mean, mode, present."); declareOption(ol, "merge_path", &MergeDond2Files::merge_path, OptionBase::buildoption, "The root name of merge files to be created.\n" "3 files will be created: a root_train.pmat, a root_test.pmat and a root_uknown.pmat"); declareOption(ol, "sec_key", &MergeDond2Files::sec_key, OptionBase::buildoption, "The column of the merge key in the secondary dataset."); declareOption(ol, "main_key", &MergeDond2Files::main_key, OptionBase::buildoption, "The column of the merge key in the main dataset."); declareOption(ol, "train_ind", &MergeDond2Files::train_ind, OptionBase::buildoption, "The column of the indicator of a train record in the main dataset."); declareOption(ol, "test_ind", &MergeDond2Files::test_ind, OptionBase::buildoption, "The column of the indicator of a test record in the main dataset."); declareOption(ol, "train_file", &MergeDond2Files::train_file, OptionBase::learntoption, "The train file created."); declareOption(ol, "test_file", &MergeDond2Files::test_file, OptionBase::learntoption, "The test file created."); declareOption(ol, "unknown_file", &MergeDond2Files::unknown_file, OptionBase::learntoption, "The unknown target file created."); inherited::declareOptions(ol); }
static const PPath& PLearn::MergeDond2Files::declaringFile | ( | ) | [inline, static] |
Reimplemented from PLearn::PLearner.
Definition at line 119 of file MergeDond2Files.h.
:
//##### Protected Member Functions ######################################
MergeDond2Files * PLearn::MergeDond2Files::deepCopy | ( | CopiesMap & | copies | ) | const [virtual] |
Reimplemented from PLearn::PLearner.
Definition at line 53 of file MergeDond2Files.cc.
OptionList & PLearn::MergeDond2Files::getOptionList | ( | ) | const [virtual] |
Reimplemented from PLearn::Object.
Definition at line 53 of file MergeDond2Files.cc.
OptionMap & PLearn::MergeDond2Files::getOptionMap | ( | ) | const [virtual] |
Reimplemented from PLearn::Object.
Definition at line 53 of file MergeDond2Files.cc.
RemoteMethodMap & PLearn::MergeDond2Files::getRemoteMethodMap | ( | ) | const [virtual] |
Reimplemented from PLearn::Object.
Definition at line 53 of file MergeDond2Files.cc.
TVec< string > PLearn::MergeDond2Files::getTestCostNames | ( | ) | const [virtual] |
*** SUBCLASS WRITING: ***
This should return the names of the costs computed by computeCostsFromOutputs.
Implements PLearn::PLearner.
Definition at line 448 of file MergeDond2Files.cc.
References PLearn::TVec< T >::append().
{ TVec<string> result; result.append( "MSE" ); return result; }
TVec< string > PLearn::MergeDond2Files::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 454 of file MergeDond2Files.cc.
References PLearn::TVec< T >::append().
{ TVec<string> result; result.append( "MSE" ); return result; }
void PLearn::MergeDond2Files::makeDeepCopyFromShallowCopy | ( | CopiesMap & | copies | ) | [virtual] |
Transforms a shallow copy into a deep copy.
Reimplemented from PLearn::PLearner.
Definition at line 122 of file MergeDond2Files.cc.
References PLearn::deepCopyField().
{ deepCopyField(external_dataset, copies); deepCopyField(missing_instructions, copies); deepCopyField(merge_instructions, copies); deepCopyField(merge_path, copies); deepCopyField(sec_key, copies); deepCopyField(main_key, copies); deepCopyField(train_ind, copies); deepCopyField(test_ind, copies); deepCopyField(train_file, copies); deepCopyField(test_file, copies); deepCopyField(unknown_file, copies); inherited::makeDeepCopyFromShallowCopy(copies); }
void PLearn::MergeDond2Files::mergeFiles | ( | ) | [private] |
Definition at line 162 of file MergeDond2Files.cc.
References PLERROR, and PLearn::ProgressBar::update().
{ // initialization with merge instructions sec_row = 0; sec_length = external_dataset->length(); sec_width = external_dataset->width(); sec_names.resize(sec_width); sec_ins.resize(sec_width); sec_input.resize(sec_width); extension_width = 0; sec_names << external_dataset->fieldNames(); for (int sec_col = 0; sec_col < sec_width; sec_col++) { sec_ins[sec_col] = "mean"; } for (int ins_col = 0; ins_col < merge_instructions.size(); ins_col++) { int sec_col; for (sec_col = 0; sec_col < sec_width; sec_col++) { if (merge_instructions[ins_col].first == sec_names[sec_col]) break; } if (sec_col >= sec_width) PLERROR("In MergeDond2Files: no field with this name in external_dataset data set: %s", (merge_instructions[ins_col].first).c_str()); if (merge_instructions[ins_col].second == "skip") sec_ins[sec_col] = "skip"; else if (merge_instructions[ins_col].second == "mean") sec_ins[sec_col] = "mean"; else if (merge_instructions[ins_col].second == "mode") sec_ins[sec_col] = "mode"; else if (merge_instructions[ins_col].second == "present") sec_ins[sec_col] = "present"; else PLERROR("In MergeDond2Files: unsupported merge instruction: %s", (merge_instructions[ins_col].second).c_str()); if (sec_ins[sec_col] != "skip") extension_width += 1; } extension_pos.resize(sec_width); extension_names.resize(extension_width); for (int ext_col = 0, sec_col = 0; sec_col < sec_width; sec_col++) { if (sec_ins[sec_col] == "skip") { extension_pos[sec_col] = -1; } else { extension_pos[sec_col] = ext_col; extension_names[ext_col] = sec_names[sec_col]; ext_col += 1; } } sec_values.resize(extension_width, 10); sec_value_cnt.resize(extension_width, 10); sec_value_ind.resize(extension_width); sec_values.clear(); sec_value_cnt.clear(); sec_value_ind.clear(); external_dataset->getRow(sec_row, sec_input); // initialize primary dataset main_row = 0; main_length = train_set->length(); main_width = train_set->width(); main_input.resize(main_width); main_names.resize(main_width); main_ins.resize(main_width); main_names << train_set->fieldNames(); primary_width = 0; for (int main_col = 0; main_col < main_width; main_col++) { main_ins[main_col] = "as_is"; } for (int ins_col = 0; ins_col < missing_instructions.size(); ins_col++) { int main_col = 0; for (main_col = 0; main_col < main_width; main_col++) { if (missing_instructions[ins_col].first == main_names[main_col]) break; } if (main_col >= main_width) PLERROR("In MergeDond2Files: no field with this name in external_dataset data set: %s", (missing_instructions[ins_col].first).c_str()); if (missing_instructions[ins_col].second == "skip") main_ins[main_col] = "skip"; else if (missing_instructions[ins_col].second == "as_is") main_ins[main_col] = "as_is"; else if (missing_instructions[ins_col].second == "zero_is_missing") main_ins[main_col] = "zero_is_missing"; else if (missing_instructions[ins_col].second == "2436935_is_missing") main_ins[main_col] = "2436935_is_missing"; else if (missing_instructions[ins_col].second == "present") main_ins[main_col] = "present"; else PLERROR("In MergeDond2Files: unsupported merge instruction: %s", (missing_instructions[ins_col].second).c_str()); if (main_ins[main_col] != "skip") primary_width += 1; } primary_names.resize(primary_width); for (int main_col = 0, prim_col = 0; main_col < main_width; main_col++) { if (main_ins[main_col] != "skip") { primary_names[prim_col] = main_names[main_col]; prim_col += 1; } } // initialize output datasets merge_col = 0; merge_width = primary_width + extension_width; merge_output.resize(merge_width); merge_names.resize(merge_width); for (int prim_col = 0; prim_col < primary_width; prim_col++) { merge_names[merge_col] = primary_names[prim_col]; merge_col +=1; } for (int ext_col = 0; ext_col < extension_width; ext_col++) { merge_names[merge_col] = extension_names[ext_col]; merge_col +=1; } train_length = 0; test_length = 0; unknown_length = 0; ProgressBar* pb = 0; pb = new ProgressBar( "Counting the number of records in the train, test and unknown datasets", main_length); for (main_row = 0; main_row < main_length; main_row++) { train_set->getRow(main_row, main_input); if (main_input[train_ind] > 0.0) train_length += 1; else if (main_input[test_ind] > 0.0) test_length += 1; else unknown_length += 1; pb->update( main_row ); } delete pb; train_file = new FileVMatrix(merge_path + "_train.pmat", train_length, merge_width); test_file = new FileVMatrix(merge_path + "_test.pmat", test_length, merge_width); unknown_file = new FileVMatrix(merge_path + "_unknown.pmat", unknown_length, merge_width); train_file->declareFieldNames(merge_names); test_file->declareFieldNames(merge_names); unknown_file->declareFieldNames(merge_names); train_row = 0; test_row = 0; unknown_row = 0; //Now, we can merge pb = new ProgressBar( "Merging primary and secondary datasets with instructions", main_length); for (main_row = 0; main_row < main_length; main_row++) { train_set->getRow(main_row, main_input); if (sec_row >= sec_length) { combineAndPut(); continue; } while (sec_input[sec_key] < main_input[main_key]) { sec_row += 1; if (sec_row >= sec_length) break; external_dataset->getRow(sec_row, sec_input); } while (sec_input[sec_key] == main_input[main_key]) { accumulateVec(); sec_row += 1; if (sec_row >= sec_length) break; external_dataset->getRow(sec_row, sec_input); } combineAndPut(); pb->update( main_row ); } delete pb; }
int PLearn::MergeDond2Files::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 444 of file MergeDond2Files.cc.
{return 0;}
void PLearn::MergeDond2Files::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 445 of file MergeDond2Files.cc.
{}
Reimplemented from PLearn::PLearner.
Definition at line 119 of file MergeDond2Files.h.
int PLearn::MergeDond2Files::ext_col [private] |
Definition at line 155 of file MergeDond2Files.h.
TVec<string> PLearn::MergeDond2Files::extension_names [private] |
Definition at line 157 of file MergeDond2Files.h.
TVec<int> PLearn::MergeDond2Files::extension_pos [private] |
Definition at line 156 of file MergeDond2Files.h.
int PLearn::MergeDond2Files::extension_width [private] |
Definition at line 154 of file MergeDond2Files.h.
### declare public option fields (such as build options) here Start your comments with Doxygen-compatible comments such as //!
The secondary dataset to merge with the main one. The main one is provided as the train_set.
Definition at line 65 of file MergeDond2Files.h.
Referenced by declareOptions().
Vec PLearn::MergeDond2Files::main_input [private] |
Definition at line 168 of file MergeDond2Files.h.
TVec<string> PLearn::MergeDond2Files::main_ins [private] |
Definition at line 170 of file MergeDond2Files.h.
The column of the merge key in the main dataset.
Definition at line 82 of file MergeDond2Files.h.
Referenced by declareOptions().
int PLearn::MergeDond2Files::main_length [private] |
Definition at line 165 of file MergeDond2Files.h.
TVec<string> PLearn::MergeDond2Files::main_names [private] |
Definition at line 169 of file MergeDond2Files.h.
int PLearn::MergeDond2Files::main_row [private] |
Definition at line 167 of file MergeDond2Files.h.
int PLearn::MergeDond2Files::main_width [private] |
Definition at line 166 of file MergeDond2Files.h.
int PLearn::MergeDond2Files::merge_col [private] |
Definition at line 182 of file MergeDond2Files.h.
TVec< pair<string, string> > PLearn::MergeDond2Files::merge_instructions |
The variable merge instructions in the form of pairs field : instruction.
Supported instructions are skip, mean, mode, present.
Definition at line 73 of file MergeDond2Files.h.
Referenced by declareOptions().
TVec<string> PLearn::MergeDond2Files::merge_names [private] |
Definition at line 184 of file MergeDond2Files.h.
Vec PLearn::MergeDond2Files::merge_output [private] |
Definition at line 183 of file MergeDond2Files.h.
The file name of the merge file to be created.
Definition at line 76 of file MergeDond2Files.h.
Referenced by declareOptions().
int PLearn::MergeDond2Files::merge_width [private] |
Definition at line 178 of file MergeDond2Files.h.
TVec< pair<string, string> > PLearn::MergeDond2Files::missing_instructions |
The variable missing regeneration instructions in the form of pairs field : instruction.
Supported instructions are skip, as_is, zero_is_missing, 2436935_is_missing.
Definition at line 69 of file MergeDond2Files.h.
Referenced by declareOptions().
TVec<string> PLearn::MergeDond2Files::primary_names [private] |
Definition at line 172 of file MergeDond2Files.h.
int PLearn::MergeDond2Files::primary_width [private] |
Definition at line 171 of file MergeDond2Files.h.
Vec PLearn::MergeDond2Files::sec_input [private] |
Definition at line 151 of file MergeDond2Files.h.
TVec<string> PLearn::MergeDond2Files::sec_ins [private] |
Definition at line 153 of file MergeDond2Files.h.
The column of the merge key in the secondary dataset.
Definition at line 79 of file MergeDond2Files.h.
Referenced by declareOptions().
int PLearn::MergeDond2Files::sec_length [private] |
Definition at line 148 of file MergeDond2Files.h.
TVec<string> PLearn::MergeDond2Files::sec_names [private] |
Definition at line 152 of file MergeDond2Files.h.
int PLearn::MergeDond2Files::sec_row [private] |
Definition at line 150 of file MergeDond2Files.h.
Mat PLearn::MergeDond2Files::sec_value_cnt [private] |
Definition at line 159 of file MergeDond2Files.h.
Definition at line 162 of file MergeDond2Files.h.
bool PLearn::MergeDond2Files::sec_value_found [private] |
Definition at line 161 of file MergeDond2Files.h.
TVec<int> PLearn::MergeDond2Files::sec_value_ind [private] |
Definition at line 160 of file MergeDond2Files.h.
Mat PLearn::MergeDond2Files::sec_values [private] |
Definition at line 158 of file MergeDond2Files.h.
int PLearn::MergeDond2Files::sec_width [private] |
Definition at line 149 of file MergeDond2Files.h.
The test file created.
Definition at line 94 of file MergeDond2Files.h.
Referenced by declareOptions().
The column of the indicator of a test record in the main dataset.
Definition at line 88 of file MergeDond2Files.h.
Referenced by declareOptions().
int PLearn::MergeDond2Files::test_length [private] |
Definition at line 176 of file MergeDond2Files.h.
int PLearn::MergeDond2Files::test_row [private] |
Definition at line 180 of file MergeDond2Files.h.
The train file created.
Definition at line 91 of file MergeDond2Files.h.
Referenced by declareOptions().
The column of the indicator of a train record in the main dataset.
Definition at line 85 of file MergeDond2Files.h.
Referenced by declareOptions().
int PLearn::MergeDond2Files::train_length [private] |
Definition at line 175 of file MergeDond2Files.h.
int PLearn::MergeDond2Files::train_row [private] |
Definition at line 179 of file MergeDond2Files.h.
The unknown target file created.
Definition at line 97 of file MergeDond2Files.h.
Referenced by declareOptions().
int PLearn::MergeDond2Files::unknown_length [private] |
Definition at line 177 of file MergeDond2Files.h.
int PLearn::MergeDond2Files::unknown_row [private] |
Definition at line 181 of file MergeDond2Files.h.