code:feature_selection
Differences
This shows you the differences between two versions of the page.
| Both sides previous revision Previous revision | |||
|
code:feature_selection [2016/08/09 10:25] 127.0.0.1 external edit |
code:feature_selection [2016/11/01 14:29] (current) asa |
||
|---|---|---|---|
| Line 5: | Line 5: | ||
| <file python feature_selection.py> | <file python feature_selection.py> | ||
| + | |||
| """ | """ | ||
| ================================================= | ================================================= | ||
| Line 30: | Line 31: | ||
| # for the features (any linear classifier will work): | # for the features (any linear classifier will work): | ||
| classifier = LinearSVC() | classifier = LinearSVC() | ||
| - | selector = RFE(classifier, step=0.1,n_features_to_select=25) | + | selector = RFE(classifier, step=0.1, n_features_to_select=25) |
| # run feature selection: | # run feature selection: | ||
| selector = selector.fit(X, y) | selector = selector.fit(X, y) | ||
| Line 43: | Line 44: | ||
| # the wrong way to perform cross-validation: | # the wrong way to perform cross-validation: | ||
| cv = cross_validation.StratifiedKFold(y, 5, shuffle=True, random_state=0) | cv = cross_validation.StratifiedKFold(y, 5, shuffle=True, random_state=0) | ||
| - | print np.mean(cross_validation.cross_val_score(classifier, Xt, y, cv=cv)) | + | print (np.mean(cross_validation.cross_val_score(classifier, Xt, y, cv=cv))) |
| # now let's perform nested cross-validation: | # now let's perform nested cross-validation: | ||
| Line 50: | Line 51: | ||
| rfe_svm = make_pipeline(selector, classifier) | rfe_svm = make_pipeline(selector, classifier) | ||
| - | print np.mean(cross_validation.cross_val_score(rfe_svm, X, y, cv=cv)) | + | print (np.mean(cross_validation.cross_val_score(rfe_svm, X, y, cv=cv))) |
| # feature selection using a univariate filter method: | # feature selection using a univariate filter method: | ||
| Line 58: | Line 59: | ||
| filter_svm = make_pipeline(filter_selector, classifier) | filter_svm = make_pipeline(filter_selector, classifier) | ||
| - | print np.mean(cross_validation.cross_val_score(filter_svm, X, y, cv=cv)) | + | print (np.mean(cross_validation.cross_val_score(filter_svm, X, y, cv=cv))) |
| </file> | </file> | ||
code/feature_selection.txt ยท Last modified: 2016/11/01 14:29 by asa
Page Tools
Except where otherwise noted, content on this wiki is licensed under the following license: CC Attribution-Share Alike 4.0 International
