def fit(self,
data_train,
labels_train,
verbose=False,
unit_test=False,
max_iter=-1):
param_grid = {
'max_depth': np.arange(3, 10),
'n_estimators': [10, 50, 100, 200]
}
if unit_test:
param_grid = [{'max_depth': [3], 'n_estimators': [10]}]
rf = RandomForestClassifier(max_depth=2, random_state=0)
self.clf = GridSearchCV(rf, param_grid, verbose=verbose)
self.clf.fit(data_train, labels_train)
self.accuracy_tuple_train = compute_accuracy_tuple(
labels_train, self.clf.predict(data_train))
def fit(self,
data_train,
labels_train,
verbose=False,
unit_test=False,
max_iter=-1):
param_grid = [{
'C': [1, 10, 100, 1000],
'kernel': ['linear']
}, {
'C': [1, 10, 100, 1000],
'gamma': [0.001, 0.0001],
'kernel': ['rbf']
}]
if unit_test:
param_grid = [{'C': [1], 'kernel': ['linear']}]
svc = SVC_(max_iter=max_iter)
self.clf = GridSearchCV(svc, param_grid, verbose=verbose)
self.clf.fit(data_train, labels_train)
self.accuracy_tuple_train = compute_accuracy_tuple(
labels_train, self.clf.predict(data_train))
return self.clf.score(data_train, labels_train)
def score(self, data_test, labels_test):
labels_pred = self.predict_scmap_cluster(data_test, labels_test)
self.test_tuple = compute_accuracy_tuple(labels_test, labels_pred)
return np.mean(labels_pred == labels_test)
def score(self, data_test, labels_test):
self.accuracy_tuple_test = compute_accuracy_tuple(
labels_test, self.clf.predict(data_test))
return self.clf.score(data_test, labels_test)