|
|
@@ -17,6 +17,33 @@ def predict(input_data: pd.DataFrame, model_name):
|
|
|
return predictions.tolist()
|
|
|
|
|
|
|
|
|
+def train_and_save_model(dataset_id, model_type, model_name, model_description):
|
|
|
+ dataset = get_dataset_by_id(dataset_id)
|
|
|
+ if dataset.empty:
|
|
|
+ raise ValueError(f"Dataset {dataset_id} is empty or not found.")
|
|
|
+
|
|
|
+ # Step 1: 数据准备
|
|
|
+ X = dataset.iloc[:, :-1] # 特征数据
|
|
|
+ y = dataset.iloc[:, -1] # 目标变量
|
|
|
+
|
|
|
+ # Step 2: 训练模型
|
|
|
+ model = train_model_by_type(X, y, model_type)
|
|
|
+
|
|
|
+ # Step 3: 保存模型到数据库
|
|
|
+ # 使用提供的 model_name 和 model_description
|
|
|
+ saved_model = save_model(model_name, model_type, model_description)
|
|
|
+
|
|
|
+ # Step 4: 保存模型参数
|
|
|
+ save_model_parameters(model, saved_model.ModelID)
|
|
|
+
|
|
|
+ # Step 5: 计算评估指标（比如MSE）
|
|
|
+ y_pred = model.predict(X)
|
|
|
+ mse = mean_squared_error(y, y_pred)
|
|
|
+
|
|
|
+ return saved_model, mse
|
|
|
+
|
|
|
+
|
|
|
+
|
|
|
if __name__ == '__main__':
|
|
|
# 反酸模型预测
|
|
|
# 测试 predict 函数
|
