05221836

app/logs/2025-05-22/south-forecast.2025-05-22.0.log

@@ -1320,3 +1320,632 @@
 2025-05-22 16:25:18,562 - dbmg.py - INFO - ✅ 缩放器 lstm 保存成功 | 文档ID: 682edf6e4f4d66c1538080fe - insert_scaler_model_into_mongo
 2025-05-22 16:25:18,564 - dbmg.py - INFO - ✅ 模型 lstm 保存成功 | 文档ID: 682edf6ee396be238e929ca2 - insert_trained_model_into_mongo
 2025-05-22 16:25:18,576 - dbmg.py - INFO - ✅ 缩放器 lstm 保存成功 | 文档ID: 682edf6ee396be238e929ca4 - insert_scaler_model_into_mongo
+2025-05-22 16:36:42,098 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 16:37:06,300 - main.py - INFO - 输入文件目录: E:/compete/app/model/data/DQYC/qy/62/1002/2025-04-21/IN - main
+2025-05-22 16:37:29,931 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 16:37:56,059 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 16:37:56,060 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 16:37:56,556 - tf_model_train.py - INFO - GPU 2 allocated - _setup_resources
+2025-05-22 16:37:56,556 - tf_model_train.py - INFO - GPU 1 allocated - _setup_resources
+2025-05-22 16:37:56,578 - data_cleaning.py - INFO - 开始清洗：训练集…… - cleaning
+2025-05-22 16:37:56,578 - data_cleaning.py - INFO - 开始清洗：训练集…… - cleaning
+2025-05-22 16:37:56,606 - data_cleaning.py - INFO - 行清洗：清洗的行数有：69，缺失的列有： - key_field_row_cleaning
+2025-05-22 16:37:56,606 - data_cleaning.py - INFO - 行清洗：清洗的行数有：68，缺失的列有： - key_field_row_cleaning
+2025-05-22 16:37:56,647 - data_handler.py - INFO - 数据总数：2907, 时序缺失的间隔：0, 其中，较长的时间间隔：0 - missing_time_splite
+2025-05-22 16:37:56,647 - data_handler.py - INFO - 数据总数：2908, 时序缺失的间隔：0, 其中，较长的时间间隔：0 - missing_time_splite
+2025-05-22 16:37:56,647 - data_handler.py - INFO - 需要补值的总点数：0 - missing_time_splite
+2025-05-22 16:37:56,649 - data_handler.py - INFO - 再次测算，需要插值的总点数为：0.0 - fill_train_data
+2025-05-22 16:37:56,649 - data_handler.py - INFO - 再次测算，需要插值的总点数为：0.0 - fill_train_data
+2025-05-22 16:37:58,554 - dbmg.py - INFO - ⚠️ 未找到模型 'lstm' 的有效记录 - get_keras_model_from_mongo
+2025-05-22 16:37:58,554 - tf_lstm.py - INFO - 加强训练加载模型权重失败:('cannot unpack non-iterable NoneType object',) - train_init
+2025-05-22 16:37:58,580 - dbmg.py - INFO - ⚠️ 未找到模型 'lstm' 的有效记录 - get_keras_model_from_mongo
+2025-05-22 16:37:58,580 - tf_lstm.py - INFO - 加强训练加载模型权重失败:('cannot unpack non-iterable NoneType object',) - train_init
+2025-05-22 16:38:35,605 - tf_lstm.py - INFO - -----模型训练经过100轮迭代----- - training
+2025-05-22 16:38:35,605 - tf_lstm.py - INFO - 训练集损失函数为：[9.0578e-01 3.1937e-01 1.0006e-01 2.7970e-02 7.1800e-03 1.9700e-03
+ 8.2000e-04 5.9000e-04 5.5000e-04 5.4000e-04 5.4000e-04 5.4000e-04
+ 5.4000e-04 5.4000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04
+ 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04
+ 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.1000e-04
+ 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04
+ 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04
+ 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04] - training
+2025-05-22 16:38:35,606 - tf_lstm.py - INFO - 验证集损失函数为：[0.51488 0.16886 0.04954 0.01328 0.00368 0.00146 0.001   0.00092 0.0009
+ 0.00089 0.00089 0.00088 0.00088 0.00088 0.00087 0.00087 0.00087 0.00087
+ 0.00086 0.00086 0.00086 0.00086 0.00086 0.00086 0.00086 0.00085 0.00085
+ 0.00085 0.00085 0.00085 0.00085 0.00085 0.00085 0.00085 0.00084 0.00084
+ 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084
+ 0.00084 0.00084 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083
+ 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083
+ 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00082 0.00082
+ 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082
+ 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082
+ 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082
+ 0.00082] - training
+2025-05-22 16:38:35,635 - tf_lstm.py - INFO - -----模型训练经过100轮迭代----- - training
+2025-05-22 16:38:35,635 - tf_lstm.py - INFO - 训练集损失函数为：[8.9900e-01 3.1653e-01 9.8870e-02 2.7350e-02 6.6900e-03 1.4900e-03
+ 3.3000e-04 1.0000e-04 6.0000e-05 6.0000e-05 6.0000e-05 6.0000e-05
+ 6.0000e-05 6.0000e-05 6.0000e-05 6.0000e-05 6.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05] - training
+2025-05-22 16:38:35,636 - tf_lstm.py - INFO - 验证集损失函数为：[5.1022e-01 1.6676e-01 4.8400e-02 1.2390e-02 2.8500e-03 6.4000e-04
+ 1.8000e-04 1.0000e-04 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05] - training
+2025-05-22 16:38:35,708 - dbmg.py - INFO - ✅ 模型 lstm 保存成功 | 文档ID: 682ee28b27b5da75067d35a8 - insert_trained_model_into_mongo
+2025-05-22 16:38:35,719 - dbmg.py - INFO - ✅ 模型 lstm 保存成功 | 文档ID: 682ee28b7a38bab92a0a5a3a - insert_trained_model_into_mongo
+2025-05-22 16:38:35,731 - dbmg.py - INFO - ✅ 缩放器 lstm 保存成功 | 文档ID: 682ee28b7a38bab92a0a5a3c - insert_scaler_model_into_mongo
+2025-05-22 16:38:35,741 - dbmg.py - INFO - ✅ 缩放器 lstm 保存成功 | 文档ID: 682ee28b27b5da75067d35aa - insert_scaler_model_into_mongo
+2025-05-22 16:41:06,720 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 16:41:06,848 - main.py - INFO - 输入文件目录: E:/compete/app/model/data/DQYC/qy/62/1002/2025-04-21/IN - main
+2025-05-22 16:41:09,615 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 16:41:36,827 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 16:41:52,435 - main.py - INFO - 输入文件目录: E:/compete/app/model/data/DQYC/qy/62/1002/2025-04-21/IN - main
+2025-05-22 16:42:15,970 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 16:42:47,697 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 16:42:47,703 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 16:42:48,195 - tf_model_train.py - INFO - GPU 1 allocated - _setup_resources
+2025-05-22 16:42:48,200 - tf_model_train.py - INFO - GPU 2 allocated - _setup_resources
+2025-05-22 16:42:48,218 - data_cleaning.py - INFO - 开始清洗：训练集…… - cleaning
+2025-05-22 16:42:48,221 - data_cleaning.py - INFO - 开始清洗：训练集…… - cleaning
+2025-05-22 16:42:48,246 - data_cleaning.py - INFO - 行清洗：清洗的行数有：68，缺失的列有： - key_field_row_cleaning
+2025-05-22 16:42:48,249 - data_cleaning.py - INFO - 行清洗：清洗的行数有：69，缺失的列有： - key_field_row_cleaning
+2025-05-22 16:42:48,289 - data_handler.py - INFO - 数据总数：2908, 时序缺失的间隔：0, 其中，较长的时间间隔：0 - missing_time_splite
+2025-05-22 16:42:48,289 - data_handler.py - INFO - 需要补值的总点数：0 - missing_time_splite
+2025-05-22 16:42:48,291 - data_handler.py - INFO - 再次测算，需要插值的总点数为：0.0 - fill_train_data
+2025-05-22 16:42:48,291 - data_handler.py - INFO - 数据总数：2907, 时序缺失的间隔：0, 其中，较长的时间间隔：0 - missing_time_splite
+2025-05-22 16:42:48,291 - data_handler.py - INFO - 需要补值的总点数：0 - missing_time_splite
+2025-05-22 16:42:48,293 - data_handler.py - INFO - 再次测算，需要插值的总点数为：0.0 - fill_train_data
+2025-05-22 16:42:50,221 - dbmg.py - INFO - ⚠️ 未找到模型 'lstm' 的有效记录 - get_keras_model_from_mongo
+2025-05-22 16:42:50,222 - tf_lstm.py - INFO - 加强训练加载模型权重失败:('cannot unpack non-iterable NoneType object',) - train_init
+2025-05-22 16:42:50,251 - dbmg.py - INFO - ⚠️ 未找到模型 'lstm' 的有效记录 - get_keras_model_from_mongo
+2025-05-22 16:42:50,251 - tf_lstm.py - INFO - 加强训练加载模型权重失败:('cannot unpack non-iterable NoneType object',) - train_init
+2025-05-22 16:43:27,237 - tf_lstm.py - INFO - -----模型训练经过100轮迭代----- - training
+2025-05-22 16:43:27,238 - tf_lstm.py - INFO - 训练集损失函数为：[8.9134e-01 3.1183e-01 9.6960e-02 2.7010e-02 6.9500e-03 1.9300e-03
+ 8.1000e-04 5.9000e-04 5.5000e-04 5.4000e-04 5.4000e-04 5.4000e-04
+ 5.4000e-04 5.4000e-04 5.4000e-04 5.4000e-04 5.3000e-04 5.3000e-04
+ 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04
+ 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04
+ 5.3000e-04 5.3000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04
+ 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04
+ 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04] - training
+2025-05-22 16:43:27,238 - tf_lstm.py - INFO - 验证集损失函数为：[0.50446 0.16402 0.04786 0.01284 0.0036  0.00145 0.00101 0.00092 0.0009
+ 0.00089 0.00089 0.00089 0.00088 0.00088 0.00088 0.00087 0.00087 0.00087
+ 0.00087 0.00087 0.00086 0.00086 0.00086 0.00086 0.00086 0.00086 0.00085
+ 0.00085 0.00085 0.00085 0.00085 0.00085 0.00085 0.00085 0.00085 0.00084
+ 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084
+ 0.00084 0.00084 0.00084 0.00084 0.00083 0.00083 0.00083 0.00083 0.00083
+ 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083
+ 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083
+ 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082
+ 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082
+ 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082
+ 0.00082] - training
+2025-05-22 16:43:27,314 - tf_lstm.py - INFO - -----模型训练经过100轮迭代----- - training
+2025-05-22 16:43:27,314 - tf_lstm.py - INFO - 训练集损失函数为：[8.9997e-01 3.1582e-01 9.7960e-02 2.6850e-02 6.5100e-03 1.4400e-03
+ 3.2000e-04 1.0000e-04 6.0000e-05 6.0000e-05 6.0000e-05 6.0000e-05
+ 6.0000e-05 6.0000e-05 6.0000e-05 6.0000e-05 6.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05] - training
+2025-05-22 16:43:27,314 - tf_lstm.py - INFO - 验证集损失函数为：[5.1007e-01 1.6575e-01 4.7690e-02 1.2090e-02 2.7600e-03 6.1000e-04
+ 1.8000e-04 1.0000e-04 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05] - training
+2025-05-22 16:43:27,336 - dbmg.py - INFO - ✅ 模型 lstm 保存成功 | 文档ID: 682ee3afa37ce5e300dffc58 - insert_trained_model_into_mongo
+2025-05-22 16:43:27,385 - dbmg.py - INFO - ✅ 缩放器 lstm 保存成功 | 文档ID: 682ee3afa37ce5e300dffc5a - insert_scaler_model_into_mongo
+2025-05-22 16:43:27,429 - dbmg.py - INFO - ✅ 模型 lstm 保存成功 | 文档ID: 682ee3afded77197474391d7 - insert_trained_model_into_mongo
+2025-05-22 16:43:27,441 - dbmg.py - INFO - ✅ 缩放器 lstm 保存成功 | 文档ID: 682ee3afded77197474391d9 - insert_scaler_model_into_mongo
+2025-05-22 16:45:43,854 - task_worker.py - ERROR - Area 1002 failed: 'Datetime' - region_task
+2025-05-22 16:46:18,805 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 16:46:21,969 - main.py - INFO - 输入文件目录: E:/compete/app/model/data/DQYC/qy/62/1002/2025-04-21/IN - main
+2025-05-22 16:46:46,365 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 16:47:14,361 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 16:47:14,373 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 16:47:15,097 - tf_model_train.py - INFO - GPU 2 allocated - _setup_resources
+2025-05-22 16:47:15,100 - tf_model_train.py - INFO - GPU 1 allocated - _setup_resources
+2025-05-22 16:47:15,135 - data_cleaning.py - INFO - 开始清洗：训练集…… - cleaning
+2025-05-22 16:47:15,136 - data_cleaning.py - INFO - 开始清洗：训练集…… - cleaning
+2025-05-22 16:47:15,169 - data_cleaning.py - INFO - 行清洗：清洗的行数有：69，缺失的列有： - key_field_row_cleaning
+2025-05-22 16:47:15,173 - data_cleaning.py - INFO - 行清洗：清洗的行数有：68，缺失的列有： - key_field_row_cleaning
+2025-05-22 16:47:15,235 - data_handler.py - INFO - 数据总数：2907, 时序缺失的间隔：0, 其中，较长的时间间隔：0 - missing_time_splite
+2025-05-22 16:47:15,235 - data_handler.py - INFO - 需要补值的总点数：0 - missing_time_splite
+2025-05-22 16:47:15,239 - data_handler.py - INFO - 再次测算，需要插值的总点数为：0.0 - fill_train_data
+2025-05-22 16:47:15,241 - data_handler.py - INFO - 数据总数：2908, 时序缺失的间隔：0, 其中，较长的时间间隔：0 - missing_time_splite
+2025-05-22 16:47:15,241 - data_handler.py - INFO - 需要补值的总点数：0 - missing_time_splite
+2025-05-22 16:47:15,244 - data_handler.py - INFO - 再次测算，需要插值的总点数为：0.0 - fill_train_data
+2025-05-22 16:47:17,702 - dbmg.py - INFO - ⚠️ 未找到模型 'lstm' 的有效记录 - get_keras_model_from_mongo
+2025-05-22 16:47:17,703 - tf_lstm.py - INFO - 加强训练加载模型权重失败:('cannot unpack non-iterable NoneType object',) - train_init
+2025-05-22 16:47:17,719 - dbmg.py - INFO - ⚠️ 未找到模型 'lstm' 的有效记录 - get_keras_model_from_mongo
+2025-05-22 16:47:17,720 - tf_lstm.py - INFO - 加强训练加载模型权重失败:('cannot unpack non-iterable NoneType object',) - train_init
+2025-05-22 16:47:59,223 - tf_lstm.py - INFO - -----模型训练经过100轮迭代----- - training
+2025-05-22 16:47:59,223 - tf_lstm.py - INFO - -----模型训练经过100轮迭代----- - training
+2025-05-22 16:47:59,223 - tf_lstm.py - INFO - 训练集损失函数为：[9.0682e-01 3.1992e-01 1.0042e-01 2.8180e-02 7.2700e-03 2.0000e-03
+ 8.3000e-04 5.9000e-04 5.5000e-04 5.4000e-04 5.4000e-04 5.4000e-04
+ 5.4000e-04 5.4000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04
+ 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04
+ 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04
+ 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04
+ 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04
+ 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04] - training
+2025-05-22 16:47:59,223 - tf_lstm.py - INFO - 训练集损失函数为：[9.0271e-01 3.1676e-01 9.8510e-02 2.7120e-02 6.6000e-03 1.4600e-03
+ 3.3000e-04 1.0000e-04 6.0000e-05 6.0000e-05 6.0000e-05 6.0000e-05
+ 6.0000e-05 6.0000e-05 6.0000e-05 6.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05] - training
+2025-05-22 16:47:59,224 - tf_lstm.py - INFO - 验证集损失函数为：[0.51566 0.1692  0.04981 0.0134  0.00373 0.00148 0.00101 0.00092 0.0009
+ 0.00089 0.00089 0.00088 0.00088 0.00088 0.00087 0.00087 0.00087 0.00087
+ 0.00086 0.00086 0.00086 0.00086 0.00086 0.00086 0.00085 0.00085 0.00085
+ 0.00085 0.00085 0.00085 0.00085 0.00085 0.00085 0.00084 0.00084 0.00084
+ 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084
+ 0.00084 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083
+ 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083
+ 0.00083 0.00083 0.00083 0.00083 0.00083 0.00082 0.00082 0.00082 0.00082
+ 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082
+ 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082
+ 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082
+ 0.00082] - training
+2025-05-22 16:47:59,224 - tf_lstm.py - INFO - 验证集损失函数为：[5.1142e-01 1.6643e-01 4.8090e-02 1.2250e-02 2.8000e-03 6.2000e-04
+ 1.8000e-04 1.0000e-04 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05] - training
+2025-05-22 16:47:59,336 - dbmg.py - INFO - ✅ 模型 lstm 保存成功 | 文档ID: 682ee4bf35eeabaeac48dbca - insert_trained_model_into_mongo
+2025-05-22 16:47:59,349 - dbmg.py - INFO - ✅ 缩放器 lstm 保存成功 | 文档ID: 682ee4bf35eeabaeac48dbcc - insert_scaler_model_into_mongo
+2025-05-22 16:47:59,372 - dbmg.py - INFO - ✅ 缩放器 lstm 保存成功 | 文档ID: 682ee4bf9324837aabe55d44 - insert_scaler_model_into_mongo
+2025-05-22 16:51:53,965 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 16:51:54,086 - main.py - INFO - 输入文件目录: E:/compete/app/model/data/DQYC/qy/62/1002/2025-04-21/IN - main
+2025-05-22 16:51:56,677 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 16:51:59,595 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 16:51:59,599 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 16:51:59,803 - tf_model_train.py - INFO - GPU 2 allocated - _setup_resources
+2025-05-22 16:51:59,821 - data_cleaning.py - INFO - 开始清洗：训练集…… - cleaning
+2025-05-22 16:51:59,829 - data_cleaning.py - INFO - 行清洗：清洗的行数有：68，缺失的列有： - key_field_row_cleaning
+2025-05-22 16:51:59,829 - data_cleaning.py - INFO - 行清洗：清洗的行数有：69，缺失的列有： - key_field_row_cleaning
+2025-05-22 16:51:59,857 - data_handler.py - INFO - 数据总数：2908, 时序缺失的间隔：0, 其中，较长的时间间隔：0 - missing_time_splite
+2025-05-22 16:51:59,857 - data_handler.py - INFO - 需要补值的总点数：0 - missing_time_splite
+2025-05-22 16:51:59,858 - data_handler.py - INFO - 数据总数：2907, 时序缺失的间隔：0, 其中，较长的时间间隔：0 - missing_time_splite
+2025-05-22 16:51:59,858 - data_handler.py - INFO - 需要补值的总点数：0 - missing_time_splite
+2025-05-22 16:51:59,858 - data_handler.py - INFO - 再次测算，需要插值的总点数为：0.0 - fill_train_data
+2025-05-22 16:51:59,859 - data_handler.py - INFO - 再次测算，需要插值的总点数为：0.0 - fill_train_data
+2025-05-22 16:52:00,807 - dbmg.py - INFO - ⚠️ 未找到模型 'lstm' 的有效记录 - get_keras_model_from_mongo
+2025-05-22 16:52:00,807 - tf_lstm.py - INFO - 加强训练加载模型权重失败:('cannot unpack non-iterable NoneType object',) - train_init
+2025-05-22 16:52:00,816 - dbmg.py - INFO - ⚠️ 未找到模型 'lstm' 的有效记录 - get_keras_model_from_mongo
+2025-05-22 16:52:00,816 - tf_lstm.py - INFO - 加强训练加载模型权重失败:('cannot unpack non-iterable NoneType object',) - train_init
+2025-05-22 16:52:13,882 - tf_lstm.py - INFO - -----模型训练经过100轮迭代----- - training
+2025-05-22 16:52:13,883 - tf_lstm.py - INFO - 训练集损失函数为：[9.0366e-01 3.1743e-01 9.8920e-02 2.7540e-02 7.0700e-03 1.9500e-03
+ 8.2000e-04 6.0000e-04 5.5000e-04 5.5000e-04 5.4000e-04 5.4000e-04
+ 5.4000e-04 5.4000e-04 5.4000e-04 5.4000e-04 5.3000e-04 5.3000e-04
+ 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04
+ 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04
+ 5.3000e-04 5.3000e-04 5.3000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04
+ 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04
+ 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04] - training
+2025-05-22 16:52:13,883 - tf_lstm.py - INFO - 验证集损失函数为：[0.51271 0.16722 0.0488  0.01307 0.00364 0.00146 0.00101 0.00092 0.0009
+ 0.0009  0.00089 0.00089 0.00088 0.00088 0.00088 0.00088 0.00087 0.00087
+ 0.00087 0.00087 0.00086 0.00086 0.00086 0.00086 0.00086 0.00086 0.00086
+ 0.00085 0.00085 0.00085 0.00085 0.00085 0.00085 0.00085 0.00085 0.00085
+ 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084
+ 0.00084 0.00084 0.00084 0.00084 0.00084 0.00083 0.00083 0.00083 0.00083
+ 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083
+ 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083
+ 0.00083 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082
+ 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082
+ 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082
+ 0.00082] - training
+2025-05-22 16:52:13,924 - dbmg.py - INFO - ✅ 模型 lstm 保存成功 | 文档ID: 682ee5bde140b01fd24f34ff - insert_trained_model_into_mongo
+2025-05-22 16:52:13,932 - dbmg.py - INFO - ✅ 缩放器 lstm 保存成功 | 文档ID: 682ee5bde140b01fd24f3501 - insert_scaler_model_into_mongo
+2025-05-22 16:52:13,956 - tf_lstm.py - INFO - -----模型训练经过100轮迭代----- - training
+2025-05-22 16:52:13,957 - tf_lstm.py - INFO - 训练集损失函数为：[8.9291e-01 3.1420e-01 9.8070e-02 2.7090e-02 6.6100e-03 1.4600e-03
+ 3.3000e-04 1.0000e-04 6.0000e-05 6.0000e-05 6.0000e-05 6.0000e-05
+ 6.0000e-05 6.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05] - training
+2025-05-22 16:52:13,957 - tf_lstm.py - INFO - 验证集损失函数为：[5.0656e-01 1.6547e-01 4.7970e-02 1.2260e-02 2.8100e-03 6.3000e-04
+ 1.8000e-04 1.0000e-04 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05] - training
+2025-05-22 16:52:13,993 - dbmg.py - INFO - ✅ 模型 lstm 保存成功 | 文档ID: 682ee5bdd1747a06d97e0ef3 - insert_trained_model_into_mongo
+2025-05-22 16:52:14,030 - dbmg.py - INFO - ✅ 缩放器 lstm 保存成功 | 文档ID: 682ee5bed1747a06d97e0ef5 - insert_scaler_model_into_mongo
+2025-05-22 16:52:15,315 - task_worker.py - ERROR - Area 1002 failed: 'Datetime' - region_task
+2025-05-22 16:54:08,472 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 16:54:12,196 - main.py - INFO - 输入文件目录: E:/compete/app/model/data/DQYC/qy/62/1002/2025-04-21/IN - main
+2025-05-22 16:54:36,162 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 16:55:09,151 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 16:55:09,153 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 16:55:09,901 - tf_model_train.py - INFO - GPU 2 allocated - _setup_resources
+2025-05-22 16:55:09,941 - data_cleaning.py - INFO - 开始清洗：训练集…… - cleaning
+2025-05-22 16:55:09,941 - data_cleaning.py - INFO - 开始清洗：训练集…… - cleaning
+2025-05-22 16:55:09,977 - data_cleaning.py - INFO - 行清洗：清洗的行数有：69，缺失的列有： - key_field_row_cleaning
+2025-05-22 16:55:10,026 - data_handler.py - INFO - 数据总数：2907, 时序缺失的间隔：0, 其中，较长的时间间隔：0 - missing_time_splite
+2025-05-22 16:55:10,027 - data_handler.py - INFO - 需要补值的总点数：0 - missing_time_splite
+2025-05-22 16:55:10,027 - data_handler.py - INFO - 数据总数：2908, 时序缺失的间隔：0, 其中，较长的时间间隔：0 - missing_time_splite
+2025-05-22 16:55:10,027 - data_handler.py - INFO - 需要补值的总点数：0 - missing_time_splite
+2025-05-22 16:55:10,029 - data_handler.py - INFO - 再次测算，需要插值的总点数为：0.0 - fill_train_data
+2025-05-22 16:55:10,029 - data_handler.py - INFO - 再次测算，需要插值的总点数为：0.0 - fill_train_data
+2025-05-22 16:55:12,772 - dbmg.py - INFO - ⚠️ 未找到模型 'lstm' 的有效记录 - get_keras_model_from_mongo
+2025-05-22 16:55:12,773 - tf_lstm.py - INFO - 加强训练加载模型权重失败:('cannot unpack non-iterable NoneType object',) - train_init
+2025-05-22 16:55:12,785 - dbmg.py - INFO - ⚠️ 未找到模型 'lstm' 的有效记录 - get_keras_model_from_mongo
+2025-05-22 16:55:12,785 - tf_lstm.py - INFO - 加强训练加载模型权重失败:('cannot unpack non-iterable NoneType object',) - train_init
+2025-05-22 16:56:01,718 - tf_lstm.py - INFO - -----模型训练经过100轮迭代----- - training
+2025-05-22 16:56:01,720 - tf_lstm.py - INFO - 训练集损失函数为：[9.0185e-01 3.1664e-01 9.8400e-02 2.7020e-02 6.5600e-03 1.4500e-03
+ 3.2000e-04 1.0000e-04 6.0000e-05 6.0000e-05 6.0000e-05 6.0000e-05
+ 6.0000e-05 6.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05] - training
+2025-05-22 16:56:01,720 - tf_lstm.py - INFO - 训练集损失函数为：[9.0446e-01 3.1836e-01 9.9350e-02 2.7650e-02 7.0800e-03 1.9400e-03
+ 8.1000e-04 5.9000e-04 5.5000e-04 5.4000e-04 5.4000e-04 5.4000e-04
+ 5.4000e-04 5.4000e-04 5.4000e-04 5.3000e-04 5.3000e-04 5.3000e-04
+ 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04
+ 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.1000e-04 5.1000e-04
+ 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04
+ 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04
+ 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04] - training
+2025-05-22 16:56:01,720 - tf_lstm.py - INFO - 验证集损失函数为：[5.1114e-01 1.6636e-01 4.7960e-02 1.2180e-02 2.7800e-03 6.2000e-04
+ 1.8000e-04 1.0000e-04 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05] - training
+2025-05-22 16:56:01,720 - tf_lstm.py - INFO - 验证集损失函数为：[0.51384 0.16792 0.04904 0.0131  0.00363 0.00145 0.001   0.00092 0.0009
+ 0.00089 0.00089 0.00088 0.00088 0.00088 0.00087 0.00087 0.00087 0.00087
+ 0.00087 0.00086 0.00086 0.00086 0.00086 0.00086 0.00086 0.00085 0.00085
+ 0.00085 0.00085 0.00085 0.00085 0.00085 0.00085 0.00085 0.00084 0.00084
+ 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084
+ 0.00084 0.00084 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083
+ 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083
+ 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00082 0.00082
+ 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082
+ 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082
+ 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082
+ 0.00082] - training
+2025-05-22 16:56:01,831 - dbmg.py - INFO - ✅ 模型 lstm 保存成功 | 文档ID: 682ee6a10749b0ee0fe21ed9 - insert_trained_model_into_mongo
+2025-05-22 16:56:01,843 - dbmg.py - INFO - ✅ 缩放器 lstm 保存成功 | 文档ID: 682ee6a10749b0ee0fe21edb - insert_scaler_model_into_mongo
+2025-05-22 16:56:01,864 - dbmg.py - INFO - ✅ 模型 lstm 保存成功 | 文档ID: 682ee6a13f317b0620e79546 - insert_trained_model_into_mongo
+2025-05-22 16:56:01,899 - dbmg.py - INFO - ✅ 缩放器 lstm 保存成功 | 文档ID: 682ee6a13f317b0620e79548 - insert_scaler_model_into_mongo
+2025-05-22 18:30:17,846 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 18:30:17,971 - main.py - INFO - 输入文件目录: E:/compete/app/model/data/DQYC/qy/62/1002/2025-04-21/IN - main
+2025-05-22 18:30:20,914 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 18:30:24,052 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 18:30:24,054 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 18:30:24,282 - tf_model_train.py - INFO - GPU 1 allocated - _setup_resources
+2025-05-22 18:30:24,299 - data_cleaning.py - INFO - 开始清洗：训练集…… - cleaning
+2025-05-22 18:30:24,300 - data_cleaning.py - INFO - 开始清洗：训练集…… - cleaning
+2025-05-22 18:30:24,308 - data_cleaning.py - INFO - 行清洗：清洗的行数有：68，缺失的列有： - key_field_row_cleaning
+2025-05-22 18:30:24,308 - data_cleaning.py - INFO - 行清洗：清洗的行数有：69，缺失的列有： - key_field_row_cleaning
+2025-05-22 18:30:24,335 - data_handler.py - INFO - 数据总数：2908, 时序缺失的间隔：0, 其中，较长的时间间隔：0 - missing_time_splite
+2025-05-22 18:30:24,335 - data_handler.py - INFO - 需要补值的总点数：0 - missing_time_splite
+2025-05-22 18:30:24,335 - data_handler.py - INFO - 数据总数：2907, 时序缺失的间隔：0, 其中，较长的时间间隔：0 - missing_time_splite
+2025-05-22 18:30:24,335 - data_handler.py - INFO - 需要补值的总点数：0 - missing_time_splite
+2025-05-22 18:30:24,336 - data_handler.py - INFO - 再次测算，需要插值的总点数为：0.0 - fill_train_data
+2025-05-22 18:30:24,336 - data_handler.py - INFO - 再次测算，需要插值的总点数为：0.0 - fill_train_data
+2025-05-22 18:30:25,304 - dbmg.py - INFO - ⚠️ 未找到模型 'lstm' 的有效记录 - get_keras_model_from_mongo
+2025-05-22 18:30:25,304 - tf_lstm.py - INFO - 加强训练加载模型权重失败:('cannot unpack non-iterable NoneType object',) - train_init
+2025-05-22 18:30:25,304 - dbmg.py - INFO - ⚠️ 未找到模型 'lstm' 的有效记录 - get_keras_model_from_mongo
+2025-05-22 18:30:25,305 - tf_lstm.py - INFO - 加强训练加载模型权重失败:('cannot unpack non-iterable NoneType object',) - train_init
+2025-05-22 18:30:39,377 - tf_lstm.py - INFO - -----模型训练经过100轮迭代----- - training
+2025-05-22 18:30:39,377 - tf_lstm.py - INFO - -----模型训练经过100轮迭代----- - training
+2025-05-22 18:30:39,377 - tf_lstm.py - INFO - 训练集损失函数为：[9.0965e-01 3.2084e-01 1.0057e-01 2.8150e-02 7.2500e-03 1.9900e-03
+ 8.3000e-04 6.0000e-04 5.5000e-04 5.4000e-04 5.4000e-04 5.4000e-04
+ 5.4000e-04 5.4000e-04 5.4000e-04 5.4000e-04 5.3000e-04 5.3000e-04
+ 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04
+ 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04
+ 5.3000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.1000e-04
+ 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04
+ 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04
+ 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04] - training
+2025-05-22 18:30:39,377 - tf_lstm.py - INFO - 训练集损失函数为：[9.0351e-01 3.1783e-01 9.9160e-02 2.7360e-02 6.6700e-03 1.4700e-03
+ 3.3000e-04 1.0000e-04 6.0000e-05 6.0000e-05 6.0000e-05 6.0000e-05
+ 6.0000e-05 6.0000e-05 6.0000e-05 6.0000e-05 6.0000e-05 6.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05
+ 5.0000e-05 5.0000e-05 5.0000e-05 5.0000e-05] - training
+2025-05-22 18:30:39,378 - tf_lstm.py - INFO - 验证集损失函数为：[0.51728 0.16966 0.04982 0.01337 0.00372 0.00148 0.00101 0.00092 0.0009
+ 0.00089 0.00089 0.00089 0.00088 0.00088 0.00088 0.00087 0.00087 0.00087
+ 0.00087 0.00087 0.00086 0.00086 0.00086 0.00086 0.00086 0.00086 0.00085
+ 0.00085 0.00085 0.00085 0.00085 0.00085 0.00085 0.00085 0.00085 0.00084
+ 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084 0.00084
+ 0.00084 0.00084 0.00084 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083
+ 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083
+ 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00083 0.00082
+ 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082
+ 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082
+ 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082 0.00082
+ 0.00082] - training
+2025-05-22 18:30:39,378 - tf_lstm.py - INFO - 验证集损失函数为：[5.1250e-01 1.6734e-01 4.8480e-02 1.2370e-02 2.8300e-03 6.3000e-04
+ 1.8000e-04 1.0000e-04 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05] - training
+2025-05-22 18:30:39,422 - dbmg.py - INFO - ✅ 模型 lstm 保存成功 | 文档ID: 682efccfd9cdd24a561b9430 - insert_trained_model_into_mongo
+2025-05-22 18:30:39,444 - dbmg.py - INFO - ✅ 模型 lstm 保存成功 | 文档ID: 682efccf11d24a1b0567e1cd - insert_trained_model_into_mongo
+2025-05-22 18:30:39,447 - dbmg.py - INFO - ✅ 缩放器 lstm 保存成功 | 文档ID: 682efccfd9cdd24a561b9432 - insert_scaler_model_into_mongo
+2025-05-22 18:30:39,452 - dbmg.py - INFO - ✅ 缩放器 lstm 保存成功 | 文档ID: 682efccf11d24a1b0567e1cf - insert_scaler_model_into_mongo
+2025-05-22 18:35:43,479 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 18:35:43,599 - main.py - INFO - 输入文件目录: E:/compete/app/model/data/DQYC/qy/62/1002/2025-04-21/IN - main
+2025-05-22 18:35:46,348 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 18:35:49,304 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 18:35:49,313 - module_wrapper.py - WARNING - From E:\compete\app\model\losses.py:10: The name tf.set_random_seed is deprecated. Please use tf.compat.v1.set_random_seed instead.
+ - _tfmw_add_deprecation_warning
+2025-05-22 18:35:49,509 - tf_model_train.py - INFO - GPU 2 allocated - _setup_resources
+2025-05-22 18:35:49,518 - tf_model_train.py - INFO - GPU 1 allocated - _setup_resources
+2025-05-22 18:35:49,527 - data_cleaning.py - INFO - 开始清洗：训练集…… - cleaning
+2025-05-22 18:35:49,535 - data_cleaning.py - INFO - 行清洗：清洗的行数有：69，缺失的列有： - key_field_row_cleaning
+2025-05-22 18:35:49,536 - data_cleaning.py - INFO - 开始清洗：训练集…… - cleaning
+2025-05-22 18:35:49,545 - data_cleaning.py - INFO - 行清洗：清洗的行数有：68，缺失的列有： - key_field_row_cleaning
+2025-05-22 18:35:49,563 - data_handler.py - INFO - 数据总数：2907, 时序缺失的间隔：0, 其中，较长的时间间隔：0 - missing_time_splite
+2025-05-22 18:35:49,563 - data_handler.py - INFO - 需要补值的总点数：0 - missing_time_splite
+2025-05-22 18:35:49,564 - data_handler.py - INFO - 再次测算，需要插值的总点数为：0.0 - fill_train_data
+2025-05-22 18:35:49,572 - data_handler.py - INFO - 数据总数：2908, 时序缺失的间隔：0, 其中，较长的时间间隔：0 - missing_time_splite
+2025-05-22 18:35:49,572 - data_handler.py - INFO - 需要补值的总点数：0 - missing_time_splite
+2025-05-22 18:35:49,573 - data_handler.py - INFO - 再次测算，需要插值的总点数为：0.0 - fill_train_data
+2025-05-22 18:35:50,543 - dbmg.py - INFO - ⚠️ 未找到模型 'lstm' 的有效记录 - get_keras_model_from_mongo
+2025-05-22 18:35:50,543 - dbmg.py - INFO - ⚠️ 未找到模型 'lstm' 的有效记录 - get_keras_model_from_mongo
+2025-05-22 18:35:50,543 - tf_lstm.py - INFO - 加强训练加载模型权重失败:('cannot unpack non-iterable NoneType object',) - train_init
+2025-05-22 18:35:50,543 - tf_lstm.py - INFO - 加强训练加载模型权重失败:('cannot unpack non-iterable NoneType object',) - train_init
+2025-05-22 18:36:05,459 - tf_lstm.py - INFO - -----模型训练经过100轮迭代----- - training
+2025-05-22 18:36:05,459 - tf_lstm.py - INFO - -----模型训练经过100轮迭代----- - training
+2025-05-22 18:36:05,460 - tf_lstm.py - INFO - 训练集损失函数为：[9.0471e-01 3.1927e-01 1.0005e-01 2.7970e-02 7.1900e-03 1.9800e-03
+ 8.3000e-04 6.0000e-04 5.5000e-04 5.5000e-04 5.4000e-04 5.4000e-04
+ 5.4000e-04 5.4000e-04 5.4000e-04 5.4000e-04 5.3000e-04 5.3000e-04
+ 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04
+ 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04 5.3000e-04
+ 5.3000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04
+ 5.2000e-04 5.2000e-04 5.2000e-04 5.2000e-04 5.1000e-04 5.1000e-04
+ 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04
+ 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04
+ 5.1000e-04 5.1000e-04 5.1000e-04 5.1000e-04] - training
+2025-05-22 18:36:05,460 - tf_lstm.py - INFO - 验证集损失函数为：[5.1764e-01 1.6935e-01 4.9040e-02 1.2470e-02 2.8400e-03 6.3000e-04
+ 1.8000e-04 1.0000e-04 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05 8.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05
+ 7.0000e-05 7.0000e-05 7.0000e-05 7.0000e-05] - training
+2025-05-22 18:36:05,514 - dbmg.py - INFO - ✅ 模型 lstm 保存成功 | 文档ID: 682efe15d416e8f832fd0f55 - insert_trained_model_into_mongo
+2025-05-22 18:36:05,514 - dbmg.py - INFO - ✅ 模型 lstm 保存成功 | 文档ID: 682efe155dc17e5973916862 - insert_trained_model_into_mongo
+2025-05-22 18:36:05,546 - dbmg.py - INFO - ✅ 缩放器 lstm 保存成功 | 文档ID: 682efe155dc17e5973916864 - insert_scaler_model_into_mongo
+2025-05-22 18:36:05,561 - dbmg.py - INFO - ✅ 缩放器 lstm 保存成功 | 文档ID: 682efe15d416e8f832fd0f57 - insert_scaler_model_into_mongo
+2025-05-22 18:36:06,898 - task_worker.py - ERROR - Area 1002 failed: 'types.SimpleNamespace' object has no attribute 'cap' - region_task

app/model/main.py

@@ -12,6 +12,8 @@ import concurrent.futures
 import types
 
 from pyexpat import features
+
+from tensorflow import add_n
 from tqdm import tqdm
 import pandas as pd
 from pathlib import Path
@@ -37,6 +39,13 @@ from multiprocessing import Manager, Lock
      2.队列中的每个场站是一个子任务，还有最终的区域级子任务
 """
 
+def add_nwp(df_obj, df):
+    if df_obj.empty:
+        df_obj = df
+    else:
+        add_cols = [col for col in df_obj.columns if col not in ['PlantID', 'PlantName', 'PlantType', 'Qbsj', 'Datetime']]
+        df_obj[add_cols] = df_obj[add_cols].add(df, fill_value=0)
+    return df_obj
 
 def main():
     # ---------------------------- 解析参数 ----------------------------
@@ -83,23 +92,28 @@ def main():
 
             # 处理完成情况
             for future in concurrent.futures.as_completed(futures):
-                result = future._result
-                if result == 'success':
-                    # 分治-汇总策略得到加权后的nwp
-                    completed += 1
-                    local = result['weights']
-                    total_cap += local['cap']
-                    weighted_nwp = weighted_nwp.add(local['nwp'], fill_value=0)
-                    weighted_nwp_h = weighted_nwp_h.add(local['nwp_h'], fill_value=0)
-                    weighted_nwp_v = weighted_nwp_v.add(local['nwp_v'], fill_value=0)
-                    weighted_nwp_v_h = weighted_nwp_v_h.add(local['nwp_v_h'], fill_value=0)
-                pbar.update(1)
-                pbar.set_postfix_str(f"Completed: {completed}/{len(all_stations)}")
+                try:
+                    result = future.result()
+                    if result['status'] == 'success':
+                        # 分治-汇总策略得到加权后的nwp
+                        completed += 1
+                        local = result['weights']
+                        total_cap += local['cap']
+                        weighted_nwp = add_nwp(weighted_nwp, local['nwp'])
+                        weighted_nwp_h = add_nwp(weighted_nwp_h, local['nwp_h'])
+                        weighted_nwp_v = add_nwp(weighted_nwp_v, local['nwp_v'])
+                        weighted_nwp_v_h = add_nwp(weighted_nwp_v_h, local['nwp_v_h'])
+                    pbar.update(1)
+                    pbar.set_postfix_str(f"Completed: {completed}/{len(all_stations)}")
+                except Exception as e:
+                    print(f"Task failed: {e}")
     # 归一化处理
-    weighted_nwp /= total_cap
-    weighted_nwp_h /= total_cap
-    weighted_nwp_v /= total_cap
-    weighted_nwp_v_h /= total_cap
+    use_cols = [col for col in weighted_nwp.columns if col not in ['PlantID', 'PlantName', 'PlantType', 'Qbsj', 'Datetime']]
+    use_cols_v = [col for col in weighted_nwp_v.columns if col not in ['PlantID', 'PlantName', 'PlantType', 'Qbsj', 'Datetime']]
+    weighted_nwp[use_cols] /= total_cap
+    weighted_nwp_h[use_cols] /= total_cap
+    weighted_nwp_v[use_cols_v] /= total_cap
+    weighted_nwp_v_h[use_cols_v] /= total_cap
     data_nwps = types.SimpleNamespace(**{'nwp': weighted_nwp, 'nwp_h': weighted_nwp_h, 'nwp_v': weighted_nwp_v, 'nwp_v_h': weighted_nwp_v_h})
     print(f"Final result: {completed} stations trained successfully")
     # 区域级功率预测训练

app/model/material.py

@@ -89,8 +89,9 @@ class MaterialLoader:
 
         def local_sum(df, weight):
             """内部函数：对DataFrame进行加权求和"""
+            weighted_df = df.copy()
             columns_to_scale = [col for col in df.columns if col not in ['PlantID', 'PlantName', 'PlantType', 'Qbsj', 'Datetime']]
-            weighted_df = df[columns_to_scale] * weight
+            weighted_df[columns_to_scale] = weighted_df[columns_to_scale] * weight
             return weighted_df, weight
 
         # 从data_objects解构对象