FeatureImputer Class Reference

Feature Imputer to complete missing values. More...

#include <FeatureProcess.h>

Inheritance diagram for FeatureImputer:

Public Member Functions
	FeatureImputer (const string &feature_name)
	FeatureImputer (const string &feature_name, string init_string)
void	addStrata (string &init_string)
void	addStrata (featureStrata &strata)
void	addStrata (string &name, float resolution, float min, float max)
int	init (map< string, string > &mapper)
	The parsed fields from init command.
void	init_defaults ()
imputeMomentTypes	getMomentType (string &entry)
void	update_req_features_vec (unordered_set< string > &out_req_features, unordered_set< string > &in_req_features)
	update sets of required as input according to set required as output to processor
virtual void	copy (FeatureProcessor *processor)
int	Learn (MedFeatures &features, unordered_set< int > &ids)
int	_apply (MedFeatures &features, unordered_set< int > &ids, bool learning)
void	check_stratas_name (MedFeatures &features, map< string, string > &strata_name_conversion)
	ADD_SERIALIZATION_FUNCS (processor_type, feature_name, resolved_feature_name, missing_value, imputerStrata, moment_type, moments, histograms, strata_sizes, default_moment, default_histogram, moment_type_vec, moments_vec, default_moment_vec, leave_missing_for_small_stratas, impute_strata_with_missing, round_to_existing_value, existing_values) void dprint(const string &pref
void	print ()
	debug and print
Public Member Functions inherited from FeatureProcessor
virtual string	select_learn_matrix (const vector< string > &matrix_tags) const
	Will be called before learn to create new version for the matrix if needed - in parallel of existing matrix.
virtual void	clear ()
void	init_defaults ()
virtual void	set_feature_name (const string &feature_name)
virtual string	get_feature_name ()
virtual void	get_feature_names (vector< string > &feature_names)
int	learn (MedFeatures &features)
	PostProcess of MedFeatures - on all ids.
int	learn (MedFeatures &features, unordered_set< int > &ids)
virtual int	_apply (MedFeatures &features, unordered_set< int > &ids)
virtual int	_conditional_apply (MedFeatures &features, unordered_set< int > &ids, unordered_set< string > &req_features, bool learning)
int	apply (MedFeatures &features, bool learning)
	PostProcess of MedFeatures - on all or a subset of the ids calls virtaul function "_apply/_conditional_apply" for the specific implementation.
int	apply (MedFeatures &features, unordered_set< string > &req_features, bool learning)
int	apply (MedFeatures &features, unordered_set< int > &ids, bool learning)
int	apply (MedFeatures &features, unordered_set< int > &ids, unordered_set< string > &req_features, bool learning)
int	apply (MedFeatures &features)
int	apply (MedFeatures &features, unordered_set< string > &req_features)
int	apply (MedFeatures &features, unordered_set< int > &ids)
int	apply (MedFeatures &features, unordered_set< int > &ids, unordered_set< string > &req_features)
virtual int	init (void *processor_params)
virtual int	filter (unordered_set< string > &features)
	Filter according to a subset of features.
string	resolve_feature_name (MedFeatures &features, string substr)
	Utility : get corresponding name in MedFeatures.
virtual bool	are_features_affected (unordered_set< string > &out_req_features)
	check if a set of features is affected by the current processor
virtual bool	is_selector ()
	allows testing if this feature processor is a selector
void *	new_polymorphic (string derived_class_name)
	for polymorphic classes that want to be able to serialize/deserialize a pointer * to the derived class given its type one needs to implement this function to return a new to the derived class given its type (as in my_type)
size_t	get_processor_size ()
size_t	processor_serialize (unsigned char *blob)
virtual void	dprint (const string &pref, int rp_flag)
Public Member Functions inherited from SerializableObject
virtual int	version () const
	Relevant for serializations.
virtual string	my_class_name () const
	For better handling of serializations it is highly recommended that each SerializableObject inheriting class will implement the next method.
virtual void	serialized_fields_name (vector< string > &field_names) const
	The names of the serialized fields.
virtual void	pre_serialization ()
virtual void	post_deserialization ()
virtual size_t	get_size ()
	Gets bytes sizes for serializations.
virtual size_t	serialize (unsigned char *blob)
	Serialiazing object to blob memory. return number ob bytes wrote to memory.
virtual size_t	deserialize (unsigned char *blob)
	Deserialiazing blob to object. returns number of bytes read.
size_t	serialize_vec (vector< unsigned char > &blob)
size_t	deserialize_vec (vector< unsigned char > &blob)
virtual size_t	serialize (vector< unsigned char > &blob)
virtual size_t	deserialize (vector< unsigned char > &blob)
virtual int	read_from_file (const string &fname)
	read and deserialize model
virtual int	write_to_file (const string &fname)
	serialize model and write to file
virtual int	read_from_file_unsafe (const string &fname)
	read and deserialize model without checking version number - unsafe read
int	init_from_string (string init_string)
	Init from string.
int	init_params_from_file (string init_file)
int	init_param_from_file (string file_str, string &param)
int	update_from_string (const string &init_string)
virtual int	update (map< string, string > &map)
	Virtual to update object from parsed fields.
virtual string	object_json () const

Data Fields
float	missing_value = MED_MAT_MISSING_VALUE
bool	verbose = true
	If true will print how many missing value were in each feature.
bool	verbose_learn = false
	If true will call print after learn.
featureSetStrata	imputerStrata
int	min_samples = 50
int	leave_missing_for_small_stratas = 0
int	impute_strata_with_missing = 0
vector< imputeMomentTypes >	moment_type_vec
vector< float >	default_moment_vec
vector< vector< float > >	moments_vec
imputeMomentTypes	moment_type = IMPUTE_MMNT_MEAN
float	default_moment
vector< float >	moments
vector< pair< float, float > >	default_histogram
vector< vector< pair< float, float > > >	histograms
vector< int >	strata_sizes
int	max_samples = 100000
	Utility : maximum number of samples to take for moments calculations.
bool	round_to_existing_value = true
vector< float >	existing_values
int	fp_flag
Data Fields inherited from FeatureProcessor
string	feature_name = "unset_feature_name"
	Feature name ( + name as appears in MedFeatures) ;.
string	resolved_feature_name
FeatureProcessorTypes	processor_type = FTR_PROCESS_LAST
int	learn_nthreads
int	clean_nthreads

Additional Inherited Members
Static Public Member Functions inherited from FeatureProcessor
static FeatureProcessor *	make_processor (string processor_name)
static FeatureProcessor *	make_processor (FeatureProcessorTypes type)
static FeatureProcessor *	make_processor (string processor_name, string params)
static FeatureProcessor *	make_processor (FeatureProcessorTypes type, string params)

Detailed Description

Feature Imputer to complete missing values.

To Use this selector specify "imputer" in the fp_type

Member Function Documentation

_apply()

int FeatureImputer::_apply ( MedFeatures &  features, unordered_set< int > &  ids, bool  learning  )

virtual

Reimplemented from FeatureProcessor.

copy()

virtual void FeatureImputer::copy ( FeatureProcessor *  processor )

inlinevirtual

Reimplemented from FeatureProcessor.

init()

int FeatureImputer::init ( map< string, string > &  mapper )

virtual

The parsed fields from init command.

        if (field == "name") feature_name = entry.second;
        else if (field == "min_samples") min_samples = med_stoi(entry.second);
        else if (field == "moment_type") {
            moment_type_vec.resize(2);
            moment_type_vec[0] = getMomentType(entry.second);
            moment_type_vec[1] = moment_type_vec[0];
        }
        else if (field == "learn_moment_type") {
            moment_type_vec.resize(2, IMPUTE_MMNT_MEAN);
            moment_type_vec[0] = getMomentType(entry.second);
        }
        else if (field == "apply_moment_type") {
            moment_type_vec.resize(2, IMPUTE_MMNT_MEAN);
            moment_type_vec[1] = getMomentType(entry.second);
        }
        else if (field == "max_samples") max_samples = med_stoi(entry.second);
        else if (field == "strata") {
            boost::split(strata, entry.second, boost::is_any_of(":"));
            for (string& stratum : strata) addStrata(stratum);
        }
        else if (field == "verbose")
            verbose = stoi(entry.second) > 0;
        else if (field == "verbose_learn")
            verbose_learn = stoi(entry.second) > 0;
        else if (field == "round_to_existing_value")
            round_to_existing_value = stoi(entry.second) > 0;
        else if (field == "leave_missing_for_small_stratas") leave_missing_for_small_stratas = med_stoi(entry.second);
        else if (field == "impute_strata_with_missing") impute_strata_with_missing = med_stoi(entry.second);
        else if (field != "names" && field != "fp_type" && field != "tag")
            MLOG("Unknown parameter \'%s\' for FeatureImputer\n", field.c_str());

[FeatureImputer::init]

[FeatureImputer::init]

Reimplemented from FeatureProcessor.

Learn()

int FeatureImputer::Learn ( MedFeatures &  features, unordered_set< int > &  ids  )

virtual

Reimplemented from FeatureProcessor.

update_req_features_vec()

void FeatureImputer::update_req_features_vec ( unordered_set< string > &  out_req_features, unordered_set< string > &  in_req_features  )

virtual

update sets of required as input according to set required as output to processor

Reimplemented from FeatureProcessor.

---

The documentation for this class was generated from the following files:

• Internal/MedProcessTools/MedProcessTools/FeatureProcess.h
• Internal/MedProcessTools/MedProcessTools/FeatureProcess.cpp