def start(self):
# multi_step_regression
multi_target_forest = MultiOutputRegressor(self.model)
train_ft_mat = self.dataset.ft_mat[self.dataset.train_index]
train_label_mat = self.dataset.label_mat[self.dataset.train_index]
valid_ft_mat = self.dataset.ft_mat[self.dataset.valid_index]
valid_label_mat = self.dataset.label_mat[self.dataset.valid_index]
test_ft_mat = self.dataset.ft_mat[self.dataset.test_index]
test_label_mat = self.dataset.label_mat[self.dataset.test_index]
multi_target_forest.fit(train_ft_mat, train_label_mat)
train_pred = multi_target_forest.predict(train_ft_mat)
valid_pred = multi_target_forest.predict(valid_ft_mat)
test_pred = multi_target_forest.predict(test_ft_mat)
print('train: ', evaluate_regression(train_pred, train_label_mat))
print('valid: ', evaluate_regression(valid_pred, valid_label_mat))
print('test: ', evaluate_regression(test_pred, test_label_mat))
# return test mse mape
mse, mae, mape = evaluate_regression(test_pred, test_label_mat)
return mse, mae, mape
def start(self):
train_ft_mat = self.dataset.ft_mat[self.dataset.train_index]
train_label_mat = self.dataset.label_mat[self.dataset.train_index]
valid_ft_mat = self.dataset.ft_mat[self.dataset.valid_index]
valid_label_mat = self.dataset.label_mat[self.dataset.valid_index]
test_ft_mat = self.dataset.ft_mat[self.dataset.test_index]
test_label_mat = self.dataset.label_mat[self.dataset.test_index]
# model = xgb.train(self.params, xgb.DMatrix(train_ft_mat, train_label_mat))
multi_target_model = MultiOutputRegressor(
xgb.XGBRegressor(objective='reg:linear'))
multi_target_model.fit(train_ft_mat, train_label_mat)
train_pred = multi_target_model.predict(train_ft_mat)
valid_pred = multi_target_model.predict(valid_ft_mat)
test_pred = multi_target_model.predict(test_ft_mat)
print('train: ', evaluate_regression(train_pred, train_label_mat))
print('valid: ', evaluate_regression(valid_pred, valid_label_mat))
print('test: ', evaluate_regression(test_pred, test_label_mat))
# return test mse mape
mse, mae, mape = evaluate_regression(test_pred, test_label_mat)
return mse, mae, mape
def test(self):
self.model.eval()
pred, adj = self.model(self.dataset.ft_mat, self.adj)
label = self.dataset.label_mat.to('cpu').detach().numpy()
pred = pred.to('cpu').detach().numpy()
train_mse, train_mae, train_mape = evaluate_regression(
pred[self.dataset.train_index], label[self.dataset.train_index])
valid_mse, valid_mae, valid_mape = evaluate_regression(
pred[self.dataset.valid_index], label[self.dataset.valid_index])
test_mse, test_mae, test_mape = evaluate_regression(
pred[self.dataset.test_index], label[self.dataset.test_index])
return train_mse, valid_mse, test_mse, train_mae, valid_mae, test_mae, \
train_mape, valid_mape, test_mape, pred
def test(self):
self.model.eval()
ft_mat = self.dataset.ft_mat
ft_mat = ft_mat.view(self.dataset.node_size, -1, 1)
pred = self.model(ft_mat).to('cpu').detach().numpy()
label = self.dataset.label_mat.to('cpu').detach().numpy()
train_mse, train_mae, train_mape = evaluate_regression(
pred[self.dataset.train_index], label[self.dataset.train_index])
valid_mse, valid_mae, valid_mape = evaluate_regression(
pred[self.dataset.valid_index], label[self.dataset.valid_index])
test_mse, test_mae, test_mape = evaluate_regression(
pred[self.dataset.test_index], label[self.dataset.test_index])
return train_mse, valid_mse, test_mse, train_mae, valid_mae, test_mae, \
train_mape, valid_mape, test_mape, pred
