def leaning_curve(self, ):
optimized_model = GradientBoostingClassifier(max_features=30)
train_sizes, train_scores, valid_scores = learning_curve(
optimized_model,
self.X_train,
self.y_train,
train_sizes=[n for n in range(50, 850, 20)],
cv=3)
plot_learning_curve(train_sizes, train_scores[:, 0], valid_scores[:,
0])
def leaning_curve(self, ):
optimized_model = KNeighborsClassifier(n_neighbors=40)
train_sizes, train_scores, valid_scores = learning_curve(
optimized_model,
self.X_train,
self.y_train,
train_sizes=[n for n in range(50, 850, 20)],
cv=3)
plot_learning_curve(train_sizes, train_scores[:, 0], valid_scores[:,
0])
def leaning_curve(self, ):
optimized_model = SVC()
train_sizes, train_scores, valid_scores = learning_curve(
optimized_model,
self.X_train,
self.y_train,
train_sizes=[n for n in range(50, 850, 20)],
cv=3)
plot_learning_curve(train_sizes, train_scores[:, 0], valid_scores[:,
0])
def leaning_curve(self, ):
optimized_model = MLPClassifier(hidden_layer_sizes=(10, 10),
alpha=0.01)
train_sizes, train_scores, valid_scores = learning_curve(
optimized_model,
self.X_train,
self.y_train,
train_sizes=[n for n in range(100, 850, 10)],
cv=3)
plot_learning_curve(train_sizes, train_scores[:, 0], valid_scores[:,
0])
def leaning_curve(self,):
optimized_model = tree.DecisionTreeClassifier(min_samples_split=200, max_depth=10)
train_sizes, train_scores, valid_scores = learning_curve(optimized_model, self.X_train, self.y_train,
train_sizes=[n for n in range(100, 850, 10)],
cv=3)
plot_learning_curve(train_sizes, train_scores[:, 0], valid_scores[:, 0])