|
Medial Code Documentation
|
A virtual class of processes on MedFeatures; E.g. More...
#include <SerializableObject/SerializableObject/SerializableObject.h>#include <MedProcessTools/MedProcessTools/MedFeatures.h>#include <MedProcessTools/MedProcessTools/MedProcessUtils.h>#include <MedProcessTools/MedProcessTools/MedValueCleaner.h>#include <MedUtils/MedUtils/MedUtils.h>#include <MedAlgo/MedAlgo/BinSplitOptimizer.h>#include <unordered_set>#include "IterativeImputer.h"Go to the source code of this file.
Data Structures | |
| class | FeatureProcessor |
| class | MultiFeatureProcessor |
| A Processor which contains a vector of simpler processors Useful for applying same cleaners on a set of features, for example. More... | |
| class | FeatureBasicOutlierCleaner |
| A simple cleaner considering each value of a certain feature separatley. More... | |
| class | FeatureNormalizer |
| Feature Normalizer. More... | |
| class | featureStrata |
| class | featureSetStrata |
| When building startas on a set of several features, we build a cartesian product of their combinations: e.g. More... | |
| class | FeatureImputer |
| Feature Imputer to complete missing values. More... | |
| class | FeatureSelector |
| Feature Selector abstract class. More... | |
| class | LassoSelector |
| Feature Selector : lasso. More... | |
| class | DgnrtFeatureRemvoer |
| Feature Selector : Remove Degenerate features. More... | |
| class | univariateSelectionParams |
| class | UnivariateFeatureSelector |
| Feature Selector : Univariate. More... | |
| class | MRMRFeatureSelector |
| Feature Selector : MRMR. More... | |
| class | TagFeatureSelector |
| TagFeatureSelector - selector which leave us only with the selected "tags" given as param (if empty do nothing) and removes removed_tags (if empty do nothing) note that you can use regex notation to specify the tags To Use this selector specify "tags_selector" in the fp_type. More... | |
| class | ImportanceFeatureSelector |
| ImportanceFeatureSelector - selector which uses feature importance method for sepcific model to rank the feature importance and select them. More... | |
| class | IterativeFeatureSelector |
| IterativeFeatureSelector - Apply bottom-up or top-down iteration for feature selection. More... | |
| class | FeatureEncoder |
| FeatureEncoder - General class for encoding features - PCA, autoencoder... More... | |
| class | FeaturePCAParams |
| PCA Parameters class. More... | |
| class | FeaturePCA |
| FeaturePCA - PCA encoder. More... | |
| class | OneHotFeatProcessor |
| OneHotFeatProcessor: More... | |
| class | GetProbFeatProcessor |
| GetProbProcessor: More... | |
| class | MultiplierProcessor |
| MultiplierProcessor: More... | |
| class | MissingIndicatorProcessor |
| FeatureMissingIndicator: creates a feature that indicates if a feature is missing or not. More... | |
| class | Binning_Wrapper |
| class | BinningFeatProcessor |
| GetProbProcessor: More... | |
Macros | |
| #define | DEFAULT_FEAT_CLNR_NTHREADS 24 |
| #define | DEF_FTR_TRIMMING_SD_NUM 7 |
| #define | DEF_FTR_REMOVING_SD_NUM 14 |
| #define | DEF_MAX_SAMPLE 1000 |
Enumerations | |
| enum | FeatureProcessorTypes { FTR_PROCESS_MULTI , FTR_PROCESS_BASIC_OUTLIER_CLEANER , FTR_PROCESS_NORMALIZER , FTR_PROCESS_IMPUTER , FTR_PROCESS_DO_CALC , FTR_PROCESS_UNIVARIATE_SELECTOR , FTR_PROCESSOR_MRMR_SELECTOR , FTR_PROCESSOR_LASSO_SELECTOR , FTR_PROCESSOR_TAGS_SELECTOR , FTR_PROCESSOR_IMPORTANCE_SELECTOR , FTR_PROCESSOR_ITERATIVE_SELECTOR , FTR_PROCESS_REMOVE_DGNRT_FTRS , FTR_PROCESS_ITERATIVE_IMPUTER , FTR_PROCESS_ENCODER_PCA , FTR_PROCESS_ONE_HOT , FTR_PROCESS_GET_PROB , FTR_PROCESS_PREDICTOR_IMPUTER , FTR_PROCESS_MULTIPLIER , FTR_PROCESS_RESAMPLE_WITH_MISSING , FTR_PROCESS_DUPLICATE , FTR_PROCESS_MISSING_INDICATOR , FTR_PROCESS_BINNING , FTR_PROCESS_LAST } |
| enum | imputeMomentTypes { IMPUTE_MMNT_MEAN , IMPUTE_MMNT_MEDIAN , IMPUTE_MMNT_COMMON , IMPUTE_MMNT_SAMPLE , IMPUTE_MMNT_LAST } |
| enum | UnivariateSelectionMethod { UNIV_SLCT_PRSN = 0 , UNIV_SLCT_MI = 1 , UNIV_SLCT_DCORR = 2 , UNIV_SLCT_LAST } |
| enum | MRMRPenaltyMethod { MRMR_MAX = 0 , MRMR_MEAN = 1 , MRMR_LAST } |
Functions | |
| FeatureProcessorTypes | feature_processor_name_to_type (const string &cleaner_name) |
| void | get_all_values (MedFeatures &features, string &signalName, unordered_set< int > &ids, vector< float > &values, int max_sample=DEF_MAX_SAMPLE) |
| void | get_all_outcomes (MedFeatures &features, unordered_set< int > &ids, vector< float > &values, int max_sample=DEF_MAX_SAMPLE) |
| void | smearBins (vector< int > &bins, int nBins, int reqNbins) |
A virtual class of processes on MedFeatures; E.g.
Cleaning
| Enumerator | |
|---|---|
| FTR_PROCESS_MULTI | "multi_processor" or "multi" to create MultiFeatureProcessor |
| FTR_PROCESS_BASIC_OUTLIER_CLEANER | "basic_outlier_cleaner" or "basic_cleaner" to create FeatureBasicOutlierCleaner |
| FTR_PROCESS_NORMALIZER | "normalizer" to create FeatureNormalizer |
| FTR_PROCESS_IMPUTER | "imputer" to create FeatureImputer |
| FTR_PROCESS_DO_CALC | "do_calc" to create DoCalcFeatProcessor |
| FTR_PROCESS_UNIVARIATE_SELECTOR | "univariate_selector" to create UnivariateFeatureSelector |
| FTR_PROCESSOR_MRMR_SELECTOR | "mrmr" or "mrmr_selector" to create MRMRFeatureSelector |
| FTR_PROCESSOR_LASSO_SELECTOR | "lasso" to create LassoSelector |
| FTR_PROCESSOR_TAGS_SELECTOR | "tags_selector" to create TagFeatureSelector |
| FTR_PROCESSOR_IMPORTANCE_SELECTOR | "importance_selector" to create ImportanceFeatureSelector |
| FTR_PROCESSOR_ITERATIVE_SELECTOR | "iterative_selector" applies bottom-up or top-down iteration for feature selection. Creates IterativeFeatureSelector |
| FTR_PROCESS_REMOVE_DGNRT_FTRS | "remove_deg" to create DgnrtFeatureRemvoer |
| FTR_PROCESS_ITERATIVE_IMPUTER | "iterative_imputer" to create IterativeImputer |
| FTR_PROCESS_ENCODER_PCA | "pca" to create FeaturePCA |
| FTR_PROCESS_ONE_HOT | "one_hot" to create OneHotFeatProcessor - make one-hot features from a given feature |
| FTR_PROCESS_GET_PROB | "get_prob" to create GetProbFeatProcessor - replace categorical feature with probability of outcome in training set |
| FTR_PROCESS_PREDICTOR_IMPUTER | "predcitor_imputer" to create PredictorImputer |
| FTR_PROCESS_MULTIPLIER | "multiplier" to create MultiplierProcessor - to multiply feature by other feature |
| FTR_PROCESS_RESAMPLE_WITH_MISSING | "resample_with_missing" to create ResampleMissingProcessor - adds missing values to learn matrix |
| FTR_PROCESS_DUPLICATE | "duplicate" to create DuplicateProcessor - duplicates samples in order to do multiple imputations. |
| FTR_PROCESS_MISSING_INDICATOR | "missing_indicator" to create MissingIndicatorProcessor - creates a feature that indicates if a feature is missing or not |
| FTR_PROCESS_BINNING | "binning" to create BinningFeatProcessor - binning with one hot on the bins |
1.9.8