修改--修改时间计划

doc/时间计划.md

@@ -20,14 +20,15 @@
 ## 4. 监控系统 (1)
     - 20250226
    - [x] 资源监控
-   - [ ] 训练监控
+   - [ ] 训练监控 --暂无等后台起来再做
-   - [ ] 告警系统
+   - [ ] 告警系统 --暂无
-   - [ ] 日志聚合
+   - [x] 日志聚合
    - 1 天
 
 ## 5. FastAPI后端服务 (2)
     - 20250221~20250224
-   - [ ] API路由设计
+    - [x] 整合系统方法
+   - [x] API路由设计
    - [ ] 请求/响应模型
    - [ ] 异步处理支持
    - [ ] API文档生成

example_data_processor.py

@@ -1,117 +0,0 @@
-from function.data_processor_date import DataProcessor
-import numpy as np
-
-# 创建处理器实例
-processor = DataProcessor()
-
-# 定义数据预处理方法
-process_methods = [
-    # 缺失值处理
-    {
-        'method_name': 'SimpleImputer',
-        'params': {
-            'strategy': 'mean',
-            'missing_values': np.nan
-        }
-    },
-    # 异常值检测
-    {
-        'method_name': 'IsolationForest',
-        'params': {
-            'contamination': 0.1,
-            'random_state': 42
-        }
-    },
-    # 数据标准化
-    {
-        'method_name': 'StandardScaler',
-        'params': {
-            'with_mean': True,
-            'with_std': True
-        }
-    }
-]
-
-# 定义特征工程方法
-feature_methods = [
-    # 类别特征编码
-    {
-        'method_name': 'OneHotEncoder',
-        'params': {
-            'sparse': False,
-            'handle_unknown': 'ignore'
-        },
-
-        # columns 要处理的列名
-        'columns': ['categorical_feature1', 'categorical_feature2']
-    },
-    # 数值特征离散化
-    {
-        'method_name': 'KBinsDiscretizer',
-        'params': {
-            'n_bins': 5,
-            'encode': 'onehot',
-            'strategy': 'uniform'
-        },
-
-        # columns 要处理的列名
-        'columns': ['numeric_feature1', 'numeric_feature2']
-    },
-    # 特征选择
-    {
-        'method_name': 'SelectKBest',
-        'params': {
-            'k': 10,
-            'score_func': 'f_classif'
-        },
-
-        # columns 要处理的列名
-        'columns': ['feature1', 'feature2', 'feature3', 'feature4']
-    },
-    # 降维
-    {
-        'method_name': 'PCA',
-        'params': {
-            'n_components': 2,
-            'random_state': 42
-        },
-
-        # columns 要处理的列名, 现并不能通过列的序号来获得列, 注意文件的列名
-        'columns': ['feature5', 'feature6', 'feature7']
-    }
-]
-
-# 数据集划分参数
-split_params = {
-    # 'test_size': 0,
-    'val_size': 0.3
-}
-
-# 处理数据集
-result = processor.process_dataset(
-    input_path='dataset/dataset_raw/breast_cancer.csv',
-    output_dir='dataset/dataset_processed',
-    process_methods=process_methods,
-    # feature_methods=feature_methods,
-    feature_methods=[],
-    split_params=split_params
-)
-
-# 打印处理结果
-print("\n数据处理结果:")
-print(f"状态: {result['status']}")
-if result['status'] == 'success':
-    print("\n处理记录:")
-    record = result['process_record']
-    print(f"输入文件: {record['input_file']}")
-    print(f"处理时间: {record['timestamp']}")
-    print("\n输出文件:")
-    for key, path in record['output_files'].items():
-        print(f"{key}: {path}")
-    
-    print("\n处理步骤:")
-    for step in record['steps']:
-        if 'shape' in step:
-            print(f"{step['step']}: 数据形状 {step['shape']}")
-else:
-    print(f"错误信息: {result['message']}") 

example_experiment_list.py

@@ -1,34 +0,0 @@
-from function.model_manager import ModelManager
-
-# 创建模型管理器实例
-manager = ModelManager()
-
-# 获取实验列表
-result = manager.get_experiments(
-    page=2,
-    page_size=10,
-    include_deleted=False
-)
-
-# 打印结果
-print("\nMLFlow实验列表:")
-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实验列表:")
-    for exp in result['experiments']:
-        print(f"\n实验ID: {exp['experiment_id']}")
-        print(f"名称: {exp['name']}")
-        print(f"存储位置: {exp['artifact_location']}")
-        print(f"状态: {exp['lifecycle_stage']}")
-        print(f"创建时间: {exp['creation_time']}")
-        print(f"最后更新: {exp['last_update_time']}")
-        print(f"运行次数: {exp['runs_count']}")
-        if exp['tags']:
-            print("标签:")
-            for tag_name, tag_value in exp['tags'].items():
-                print(f"  {tag_name}: {tag_value}")
-else:
-    print(f"错误信息: {result['message']}") 

example_get_all_dataset.py

@@ -1,5 +0,0 @@
-from function.get_all_dataset import DatasetHistory
-
-
-t = DatasetHistory()
-print(t.get_dataset())

example_method_reader_date_feature.py

@@ -1,14 +0,0 @@
-from function.method_reader_date_feature import MethodReader
-
-# 创建方法读取器实例
-reader = MethodReader()
-
-# 获取所有预处理方法
-methods = reader.get_preprocessing_methods()
-print("预处理方法列表:")
-print(methods)
-
-# 获取特定方法的详细信息
-method_details = reader.get_method_details('KBinsDiscretizer')
-print("\nKBinsDiscretizer方法详情:")
-print(method_details) 

example_method_reader_date_process.py

@@ -1,14 +0,0 @@
-from function.method_reader_date_process import MethodReader
-
-# 创建方法读取器实例
-reader = MethodReader()
-
-# 获取所有预处理方法
-methods = reader.get_preprocessing_methods()
-print("预处理方法列表:")
-print(methods)
-
-# 获取特定方法的详细信息
-method_details = reader.get_method_details('StandardScaler')
-print("\nStandardScaler方法详情:")
-print(method_details) 

example_method_reader_metric.py

@@ -1,9 +0,0 @@
-
-from function.method_reader_metric import MethodReader
-
-
-
-method_reader = MethodReader()
-
-
-print(method_reader.get_metrics())

example_method_reader_model.py

@@ -1,14 +0,0 @@
-from function.method_reader_model import MethodReader
-
-# 创建方法读取器实例
-reader = MethodReader()
-
-# 获取所有预处理方法
-methods = reader.get_models()
-print("模型列表:")
-print(methods)
-
-# 获取特定方法的详细信息
-method_details = reader.get_model_details('SVC')
-print("\nSVC方法详情:")
-print(method_details) 

example_model_delete.py

@@ -1,6 +0,0 @@
-from function.model_manager import ModelManager
-
-# 创建模型管理器实例
-manager = ModelManager()
-back = manager.delete_model('7970364d490f4e0aa0375c2db26215f3')
-print(back)

example_model_predict.py

@@ -1,8 +0,0 @@
-from function.model_manager import ModelManager
-
-model_manager = ModelManager()
-
-print(model_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"] ))

example_model_trainer.py

@@ -1,56 +0,0 @@
-from function.model_trainer import ModelTrainer
-import pandas as pd
-import numpy as np
-
-# 创建训练器实例
-trainer = ModelTrainer()
-
-# 加载数据
-train_data = pd.read_csv('/home/admin-root/haotian/MLPlatform/dataset/dataset_processed/breast_cancer_20250219_144629/train_breast_cancer_20250219_144629.csv')
-val_data = pd.read_csv('/home/admin-root/haotian/MLPlatform/dataset/dataset_processed/breast_cancer_20250219_144629/val_breast_cancer_20250219_144629.csv')
-
-# 准备特征和标签
-X_train = train_data.drop('target', axis=1)
-y_train = train_data['target']
-X_val = val_data.drop('target', axis=1)
-y_val = val_data['target']
-
-# 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 = trainer.train_model(
-        {
-            'features': X_train,
-            'labels': y_train
-        },
-        {
-            'features': X_val,
-            'labels': y_val
-        },
-        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']}")