CoTraining¶
-
class
s3l.classification.CoTraining.CoTraining(pos=1, neg=1, model1=SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0, decision_function_shape='ovr', degree=3, gamma='auto', kernel='rbf', max_iter=-1, probability=True, random_state=None, shrinking=True, tol=0.001, verbose=False), model2=SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0, decision_function_shape='ovr', degree=3, gamma='auto', kernel='rbf', max_iter=-1, probability=True, random_state=None, shrinking=True, tol=0.001, verbose=False), ind1=array([], dtype=int64), ind2=array([], dtype=int64), nepo=40, buffer_size=200)[source]¶ Bases:
s3l.base.InductiveEstimatorWOGraphCoTraining classifier
Parameters: - pos (int (default=1)) – The number of positive samples selected in each contraining iteration
- neg (int (default=1)) – The number of negative samples selected in each contraining iteration
- model (object) – [model1,model2] initializing model1 for view1 and model2 for view2
- ind1 (array-like [] (default=0)) – The column index of view1 in features X
- ind2 (array-like [] (default=0)) – The column index of view2 in features X
- nepo (int (default=40)) – The number of iteration
- buffer_size (int (default=200)) – The size of buffer
-
model¶ Two best model for view1 and view2.
Type: object list
-
fit(X, targets, labeled_idx)[source]¶ Fit a cotraining model
All the input data is provided matrix X (labeled and unlabeled) and corresponding label matrix y with a dedicated marker value for unlabeled samples.
Parameters: - feature1 (view1 array-like, shape = [n_samples, n_features]) –
- feature2 (view2 array_like, shape = [n_samples, n_features]) –
- targets (array_like, shape = [n_samples]) – label of n_labeled_samples in X.
- labeled_ind (the index of labeled data in targets.) –
- example (For) –
Returns: self
Return type: returns an instance of self.