from function.model_manager import ModelManager
import pandas as pd

# 创建模型管理器实例
manager = ModelManager()



# 获取所有预处理方法
print("--------------------------------------------获取预处理方法---------------------------------------------------")
methods = manager.get_models()
print("模型列表:")
print(methods)
print("--------------------------------------------获取预处理方法 end ---------------------------------------------------")


print("--------------------------------------------获取方法详细信息---------------------------------------------------")
# 获取特定方法的详细信息
method_details = manager.get_model_details('LinearRegression')
print("\nLinearRegression方法详情:")
print(method_details) 
print("--------------------------------------------获取方法详细信息 end ---------------------------------------------------")



print("--------------------------------------------评价指标 ---------------------------------------------------")
# 获取评价指标
print(manager.get_metrics())
print("--------------------------------------------评价指标 end ---------------------------------------------------")


print("--------------------------------------------获取所有已训练模型 ---------------------------------------------------")
# 获取所有已训练模型
result = manager.get_finished_models(
    page=1,
    page_size=10,
    experiment_name='breast_cancer_classification_3'
)

# 打印结果
print("\n已训练模型列表:")
print(f"状态: {result['status']}")
if result['status'] == 'success':
    print(f"\n总数: {result['total_count']}")
    print(f"当前页: {result['page']}")
    print(f"每页数量: {result['page_size']}")
    print("\n模型列表:")
    for model in result['models']:
        '''
            'run_id': run['run_id'],
            'experiment_id': run['experiment_id'],
        '''
        print(f"run_id", model['run_id'])
        print(f"experiment_id", model['experiment_id'])
        print(f"算法: {model['algorithm']}")
        print(f"任务类型: {model['task_type']}")
        print(f"数据集: {model['dataset']}")

        print(f"训练开始时间: {model['training_start_time']}")
        print(f"训练结束时间: {model['training_end_time']}")
        print("模型参数:")
        for k, v in model['parameters'].items():
            print(f" {k}: {v}")
        print("评估指标:")
        for metric_name, metric_value in model['metrics'].items():
            print(f"  {metric_name}: {metric_value:.4f}")
else:
    print(f"错误信息: {result['message']}") 

print("--------------------------------------------获取所有已训练模型 end ---------------------------------------------------")


print("--------------------------------------------模型训练---------------------------------------------------")


# 模型配置
model_config = {
    'algorithm': 'XGBClassifier',
    'task_type': 'classification',
    # 'dataset' : '/home/admin-root/haotian/MLPlatform/dataset/dataset_processed/breast_cancer_20250219_144629',
    'params': {
        'n_estimators': 100,
        'learning_rate': 0.1,
        'max_depth': 6,
        'random_state': 42
    }
}

# 训练模型,  删除训练实验时要删除 mlruns/.trash/  回收站里的文件
# 模型文件 直接在 mlruns/文件夹下
for i in range(3, 4):
    result = manager.train_model(
        '/home/admin-root/haotian/MLPlatform/dataset/dataset_processed/breast_cancer_20250224_170615/train_breast_cancer_20250224_170615.csv',
        '/home/admin-root/haotian/MLPlatform/dataset/dataset_processed/breast_cancer_20250224_170615/val_breast_cancer_20250224_170615.csv',
        model_config,
        f'breast_cancer_classification_{i}'
    )

# 打印结果
print("\n训练结果:")
print(f"状态: {result['status']}")
if result['status'] == 'success':
    print(f"\nMLflow运行ID: {result['run_id']}")
    print("\n评估指标:")
    for metric_name, metric_value in result['metrics'].items():
        print(f"{metric_name}: {metric_value:.4f}")
else:
    print(f"错误信息: {result['message']}") 

print("-------------------------------------------模型训练 end ---------------------------------------------------")

print("--------------------------------------------模型预测 ---------------------------------------------------")
print(manager.predict(run_id = "33939ea6d8ce4d43a268f23f7361651e",\
                            data_path="/home/admin-root/haotian/MLPlatform/dataset/dataset_processed/breast_cancer_20250219_145614/test_breast_cancer_20250219_145614.csv",\
                            output_path="predictions/pred_breast_cancer_20250219_145614.csv" ,\
                            metrics= ["accuracy", "f1", "precision", "recall"] ))

print("-------------------------------------------模型预测 end ---------------------------------------------------")