Man pages for mdkessler/kTSCR
Implements k-Top Scoring Cluster Regression

all_equalCheck vector items for equality
app_minus_test_thresh_weighted_sumWeight feature importance sum on the basis of overfitting
calc_K_from_pairsCalculate K score from pairs
choose_top_featuresChoose top features
condense_k_cv_outputCondense k cross validation output
convert_sibling_names_to_indicesConvert sibling names to indices
elder_weighted_corr_scoreCalculate weighted corr score
get_app_minus_testCalculate apparent correlation minus test correlation
get_elderCalculate weighted feature correlation score
get_pairwise_rank_matricesCalculate feature pair scores
get_siblingsDetermine elder and siblings that make cluster
get_sorted_corrs_pairwise_featuresCorrelations between each pairwise feature and y
get_top_clustersImplements kTSCR algorithm
make_feature_pair_score_matrixCombine per sample feature pair score matrices into a single...
predictPredict y from K
rmseCalculate Root Mean Squared Error
run_cross_validationRun k fold cross validation
split_sibling_indicesSplit pairwise features into numeric indices
split_sibling_namesSplit pairwise features into character vector
update_pairwise_feature_matRemove elder from pairwise feature matrix
mdkessler/kTSCR documentation built on Feb. 25, 2021, 10:31 p.m.