Cross-validation helper

class mriqc.classifier.helper.CVHelper(X=None, Y=None, load_clf=None, param_file=None, n_jobs=-1, site_label='site', rate_label=None, scorer='roc_auc', b_leaveout=False, multiclass=False, verbosity=0, split='kfold', debug=False, model='rfc', basename=None, nested_cv=False, nested_cv_kfold=False, permutation_test=0)[source]

Bases: mriqc.classifier.helper.CVHelperBase

Xtest
estimator
evaluate(scoring=None, matrix=False, save_roc=False, save_pred=False)[source]

Evaluate the internal estimator on the test data

fit()[source]

Fits the cross-validation helper

fit_full()[source]

Completes the training of the model with the examples from the left-out dataset

load(filehandler)[source]

UnPickle the estimator, adding the feature names http://scikit-learn.org/stable/modules/model_persistence.html

predict(X, thres=0.5, return_proba=True)[source]

Predict class for X. The predicted class of an input sample is a vote by the trees in the forest, weighted by their probability estimates. That is, the predicted class is the one with highest mean probability estimate across the trees.

predict_dataset(data, thres=0.5, save_pred=False, site=None)[source]
save(suffix='estimator', compress=3)[source]

Pickle the estimator, adding the feature names http://scikit-learn.org/stable/modules/model_persistence.html

setXtest(X, Y)[source]
class mriqc.classifier.helper.CVHelperBase(X, Y, param_file=None, n_jobs=-1, site_label='site', rate_label=None, rate_selection='random', scorer='roc_auc', multiclass=False, verbosity=0, debug=False)[source]

Bases: object

A base helper to build cross-validation schemes

fit()[source]
ftnames
predict(X, thres=0.5, return_proba=True)[source]
predict_dataset(data, thres=0.5, save_pred=False, site=None)[source]
rate_column