- class miml.classifier.mi.miles_classifier.MILESClassifier(sigma2=20.25)#
Bases:
object
Mapping bags to an instance based feature space, from paper
Attributes#
- classifier
Classifier used from mil library
- model
Model to use
- mapping
Mapping to be used
- sigma2float
Parameter of the classifier
References#
MILES: Multiple-Instance Learning via Embedded Instance Selection (Chen et al.) http://infolab.stanford.edu/~wangz/project/imsearch/SVM/PAMI06/chen.pdf
- fit(x_train: array, y_train: array) None #
Fit the classifier to the training data.
Parameters#
- x_trainndarray of shape (n_bags, n_instances, n_features)
Features values of bags in the training set.
- y_trainndarray (n_bags, n_instances, n_labels)
Labels of bags in the training set.