Module: features

Module: features

Summary

FeaturesSetBase(features_array[, ...])

Attributes

Module API

class FeaturesSetBase(features_array, features_names=None, target_array=None, weight_array=None, features_original_entry_ID=None)[source]

Bases: object

Attributes

features
features_N_cols
features_N_rows
features_names
features_original_entry_ID
target
target_array
weight_array

Methods

add_feature(feature, feature_name)
clean_features_bad_value(bad_value)
clean_features_nan_inf()
filter_features_rows(selected)
get_feature_by_name(feature_name)
merge(feature)
new_from_rows(selected) Creates a new FeaturesSet object from a slice of an existing FeaturesSet object
new_from_table(table[, target_array, ...])
plot_contour_scatter_plot(name_1, name_2[, ...])
plot_feature_histogram(names[, bins, ...])
remove_feature(removing_feature_names_list)
select_columns_by_ID(use_IDs_list)
sort_columns(id_array)
split_train_test(train_test_split_ratio[, ...])
add_feature(feature, feature_name)[source]
clean_features_bad_value(bad_value)[source]
clean_features_nan_inf()[source]
features
features_N_cols
features_N_rows
features_names
features_original_entry_ID
filter_features_rows(selected)[source]
get_feature_by_name(feature_name)[source]
merge(feature)[source]
new_from_rows(selected)[source]

Creates a new FeaturesSet object from a slice of an existing FeaturesSet object Args:

feature_obj: selected: either bool array or IDs array

Returns:

classmethod new_from_table(table, target_array=None, weight_array=None, target_col_name=None, target_col_num=None, original_entry_ID_col_name=None, use_col_names_list=[], skip_col_names_list=[], use_col_num_list=[], skip_col_num_list=[], regex=True, rows_IDs_use_list=[], rows_IDs_skip_list=[])[source]
plot_contour_scatter_plot(name_1, name_2, n_levels=20, ax=None, exclude_outliers=False, show_plot=False, plot_contour=True, plot_scatter=True)[source]
plot_feature_histogram(names, bins=20, normed=False, KDE=False, gridsize=100, bw='scott', ax=None, plot=False, exclude_outliers=None, save_plot=False)[source]
remove_feature(removing_feature_names_list, regex=True)[source]
select_columns_by_ID(use_IDs_list)[source]
sort_columns(id_array)[source]
split_train_test(train_test_split_ratio, sampling_array=None, stratified=True)[source]
target
target_array
weight_array
write_table(data, fp, target)[source]
write_weka_file(out_file, features, features_names, class_ids=None, class_labels=None, relation='None')[source]