A123798qw
commented
6 months ago 配置都是按照github上指示的，但不管是訓練哪個模型都是出現這樣狀況
也有試過 打开PaddleOCR下的ppocr\data\simple_dataset.py，头部增加： import sys sys.setrecursionlimit(3000) #设置递归深度 跟這個方法 代码在ppocr\data\simple_dataset.py的130行：
data = {'img_path': img_path, 'label': label} # 修改前 data = {'img_path': img_path, 'label': label.replace('\r', '')} # 修改后 ————————————————
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。原文链接：https://blog.csdn.net/HEYUDAGE/article/details/126219497
UserWangZz
allenpeng88
aaptl
xdfbb
SzYang721
请提供下述完整信息以便快速定位问题/Please provide the following information to quickly locate the problem
np.boolwas a deprecated alias for the builtinbool. To avoid this error in existing code, useboolby itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, usenp.bool_here. The aliases was originally deprecated in NumPy 1.20; for more details and guidance see the original release note at: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations. Did you mean: 'bool_'?[2024/03/13 19:48:44] ppocr ERROR: When parsing line C:\Users\user\Desktop\PaddleOCR-2.6.0\train_data\det\train\ENmotor35.jpg [{"transcription": "3-PHASE", "points": [[283, 77], [419, 77], [419, 99], [283, 99]], "difficult": false, "key_cls": "PHASE"}, {"transcription": "50HP", "points": [[175, 143], [242, 143], [242, 162], [175, 162]], "difficult": false, "key_cls": "HP"}, {"transcription": "37kW", "points": [[248, 141], [313, 141], [313, 161], [248, 161]], "difficult": false, "key_cls": "kW"}, {"transcription": "200LA", "points": [[426, 134], [511, 134], [511, 157], [426, 157]], "difficult": false, "key_cls": "FRAME"}, {"transcription": "POLES2", "points": [[94, 168], [224, 168], [224, 190], [94, 190]], "difficult": false, "key_cls": "POLE"}, {"transcription": "Hz50", "points": [[108, 187], [269, 187], [269, 212], [108, 212]], "difficult": false, "key_cls": "FREQUENCY"}, {"transcription": "380", "points": [[193, 211], [285, 211], [285, 233], [193, 233]], "difficult": false, "key_cls": "VOLTS"}, {"transcription": "70", "points": [[206, 237], [248, 237], [248, 263], [206, 263]], "difficult": false, "key_cls": "AMPS"}, {"transcription": "R.PM.2940", "points": [[80, 263], [264, 263], [264, 287], [80, 287]], "difficult": false, "key_cls": "RPM"}, {"transcription": "DATE 2011", "points": [[58, 375], [208, 375], [208, 405], [58, 405]], "difficult": false, "key_cls": "DATE"}] , error happened with msg: Traceback (most recent call last): File "C:\Users\user\Desktop\PaddleOCR-2.6.0\ppocr\data\simple_dataset.py", line 138, in getitem outs = transform(data, self.ops) File "C:\Users\user\Desktop\PaddleOCR-2.6.0\ppocr\data\imaug__init.py", line 53, in transform data = op(data) File "C:\Users\user\Desktop\PaddleOCR-2.6.0\ppocr\data\imaug\label_ops.py", line 67, in call txt_tags = np.array(txt_tags, dtype=np.bool) File "C:\Users\user\anaconda3\envs\paddle310\lib\site-packages\numpy__init.py", line 324, in getattr raise AttributeError(former_attrs[attr]) AttributeError: module 'numpy' has no attribute 'bool'.
np.boolwas a deprecated alias for the builtinbool. To avoid this error in existing code, useboolby itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, usenp.bool_here. The aliases was originally deprecated in NumPy 1.20; for more details and guidance see the original release note at: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations. Did you mean: 'bool_'?[2024/03/13 19:48:44] ppocr ERROR: When parsing line C:\Users\user\Desktop\PaddleOCR-2.6.0\train_data\det\train\ENmotor12.jpg [{"transcription": "10HP", "points": [[55, 107], [109, 107], [109, 132], [55, 132]], "difficult": false, "key_cls": "HP"}, {"transcription": "FR.132M", "points": [[113, 107], [174, 107], [174, 128], [113, 128]], "difficult": false, "key_cls": "FRAME"}, {"transcription": "3 PHASE", "points": [[154, 67], [258, 67], [258, 95], [154, 95]], "difficult": false, "key_cls": "PHASE"}, {"transcription": "4P", "points": [[60, 134], [89, 134], [89, 155], [60, 155]], "difficult": false, "key_cls": "POLE"}, {"transcription": "50", "points": [[56, 185], [80, 185], [80, 207], [56, 207]], "difficult": false, "key_cls": "FREQUENCY"}, {"transcription": "50", "points": [[53, 212], [78, 212], [78, 235], [53, 235]], "difficult": false, "key_cls": "FREQUENCY"}, {"transcription": "60", "points": [[52, 240], [77, 240], [77, 260], [52, 260]], "difficult": false, "key_cls": "FREQUENCY"}, {"transcription": "60", "points": [[53, 266], [79, 266], [79, 287], [53, 287]], "difficult": false, "key_cls": "FREQUENCY"}, {"transcription": "220", "points": [[108, 184], [147, 184], [147, 202], [108, 202]], "difficult": false, "key_cls": "VOLTS"}, {"transcription": "380", "points": [[108, 209], [148, 209], [148, 230], [108, 230]], "difficult": false, "key_cls": "VOLTS"}, {"transcription": "220", "points": [[106, 236], [146, 236], [146, 255], [106, 255]], "difficult": false, "key_cls": "VOLTS"}, {"transcription": "380", "points": [[106, 262], [146, 262], [146, 288], [106, 288]], "difficult": false, "key_cls": "VOLTS"}, {"transcription": "28", "points": [[197, 176], [236, 176], [236, 203], [197, 203]], "difficult": false, "key_cls": "AMPS"}, {"transcription": "16.1", "points": [[196, 204], [239, 204], [239, 227], [196, 227]], "difficult": false, "key_cls": "AMPS"}, {"transcription": "26.6", "points": [[193, 232], [236, 232], [236, 257], [193, 257]], "difficult": false, "key_cls": "AMPS"}, {"transcription": "15.4", "points": [[196, 260], [241, 260], [241, 281], [196, 281]], "difficult": false, "key_cls": "AMPS"}, {"transcription": "1445", "points": [[265, 177], [310, 177], [310, 196], [265, 196]], "difficult": false, "key_cls": "RPM"}, {"transcription": "1445", "points": [[268, 204], [316, 204], [316, 224], [268, 224]], "difficult": false, "key_cls": "RPM"}, {"transcription": "1740", "points": [[267, 229], [313, 229], [313, 252], [267, 252]], "difficult": false, "key_cls": "RPM"}, {"transcription": "1740", "points": [[264, 257], [315, 257], [315, 279], [264, 279]], "difficult": false, "key_cls": "RPM"}, {"transcription": "86.5", "points": [[325, 176], [370, 176], [370, 197], [325, 197]], "difficult": false, "key_cls": "EFF"}, {"transcription": "86.5", "points": [[328, 202], [368, 202], [368, 224], [328, 224]], "difficult": false, "key_cls": "EFF"}, {"transcription": "87.0", "points": [[330, 226], [375, 226], [375, 250], [330, 250]], "difficult": false, "key_cls": "EFF"}, {"transcription": "87.0", "points": [[325, 257], [374, 257], [374, 280], [325, 280]], "difficult": false, "key_cls": "EFF"}] , error happened with msg: Traceback (most recent call last): File "C:\Users\user\Desktop\PaddleOCR-2.6.0\ppocr\data\simple_dataset.py", line 138, in getitem outs = transform(data, self.ops) File "C:\Users\user\Desktop\PaddleOCR-2.6.0\ppocr\data\imaug__init.py", line 53, in transform data = op(data) File "C:\Users\user\Desktop\PaddleOCR-2.6.0\ppocr\data\imaug\label_ops.py", line 67, in call txt_tags = np.array(txt_tags, dtype=np.bool) File "C:\Users\user\anaconda3\envs\paddle310\lib\site-packages\numpy__init.py", line 324, in getattr raise AttributeError(former_attrs[attr]) AttributeError: module 'numpy' has no attribute 'bool'.
np.boolwas a deprecated alias for the builtinbool. To avoid this error in existing code, useboolby itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, usenp.bool_here. The aliases was originally deprecated in NumPy 1.20; for more details and guidance see the original release note at: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations. Did you mean: 'bool_'?[2024/03/13 19:48:44] ppocr ERROR: When parsing line C:\Users\user\Desktop\PaddleOCR-2.6.0\train_data\det\train\ENmotor19.jpg [{"transcription": "3HP", "points": [[434, 278], [580, 278], [580, 327], [434, 327]], "difficult": false, "key_cls": "HP"}, {"transcription": "2.2kW", "points": [[622, 280], [800, 280], [800, 328], [622, 328]], "difficult": false, "key_cls": "kW"}, {"transcription": "182T", "points": [[990, 202], [1097, 202], [1097, 239], [990, 239]], "difficult": false, "key_cls": "FRAME"}, {"transcription": "4", "points": [[432, 323], [502, 323], [502, 363], [432, 363]], "difficult": false, "key_cls": "POLE"}, {"transcription": "3", "points": [[748, 323], [798, 323], [798, 364], [748, 364]], "difficult": false, "key_cls": "PHASE"}, {"transcription": "60", "points": [[544, 373], [629, 373], [629, 409], [544, 409]], "difficult": false, "key_cls": "FREQUENCY"}, {"transcription": "208-230/460", "points": [[450, 413], [743, 413], [743, 466], [450, 466]], "difficult": false, "key_cls": "VOLTS"}, {"transcription": "8.6-7.8/3.9", "points": [[459, 469], [694, 469], [694, 521], [459, 521]], "difficult": false, "key_cls": "AMPS"}, {"transcription": "1760／", "points": [[513, 573], [666, 573], [666, 621], [513, 621]], "difficult": false, "key_cls": "RPM"}, {"transcription": "200V", "points": [[394, 796], [536, 796], [536, 849], [394, 849]], "difficult": false, "key_cls": "VOLTS"}, {"transcription": "87.5", "points": [[1350, 735], [1450, 735], [1450, 788], [1350, 788]], "difficult": false, "key_cls": "EFF"}, {"transcription": "85.5", "points": [[1355, 791], [1451, 791], [1451, 841], [1355, 841]], "difficult": false, "key_cls": "EFF"}] , error happened with msg: Traceback (most recent call last): File "C:\Users\user\Desktop\PaddleOCR-2.6.0\ppocr\data\simple_dataset.py", line 138, in getitem outs = transform(data, self.ops) File "C:\Users\user\Desktop\PaddleOCR-2.6.0\ppocr\data\imaug__init.py", line 53, in transform data = op(data) File "C:\Users\user\Desktop\PaddleOCR-2.6.0\ppocr\data\imaug\label_ops.py", line 67, in call txt_tags = np.array(txt_tags, dtype=np.bool) File "C:\Users\user\anaconda3\envs\paddle310\lib\site-packages\numpy__init.py", line 324, in getattr raise AttributeError(former_attrs[attr]) AttributeError: module 'numpy' has no attribute 'bool'.
np.boolwas a deprecated alias for the builtinbool. To avoid this error in existing code, useboolby itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, usenp.bool_here. The aliases was originally deprecated in NumPy 1.20; for more details and guidance see the original release note at: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations. Did you mean: 'bool_'?[2024/03/13 19:48:44] ppocr ERROR: When parsing line C:\Users\user\Desktop\PaddleOCR-2.6.0\train_data\det\train\ENmotor17.jpg [{"transcription": "1.5kW", "points": [[137, 94], [242, 94], [242, 122], [137, 122]], "difficult": false, "key_cls": "kW"}, {"transcription": "4P", "points": [[272, 100], [309, 100], [309, 124], [272, 124]], "difficult": false, "key_cls": "POLE"}, {"transcription": "90L", "points": [[215, 145], [253, 145], [253, 166], [215, 166]], "difficult": false, "key_cls": "FRAME"}, {"transcription": "1440", "points": [[216, 167], [268, 167], [268, 186], [216, 186]], "difficult": false, "key_cls": "RPM"}, {"transcription": "8.7/3.9", "points": [[217, 189], [287, 189], [287, 210], [217, 210]], "difficult": false, "key_cls": "AMPS"}, {"transcription": "220/380V", "points": [[338, 97], [465, 97], [465, 124], [338, 124]], "difficult": false, "key_cls": "VOLTS"}, {"transcription": "50Hz", "points": [[495, 100], [560, 100], [560, 125], [495, 125]], "difficult": false, "key_cls": "FREQUENCY"}, {"transcription": "THREE PHASE", "points": [[374, 64], [510, 64], [510, 83], [374, 83]], "difficult": false, "key_cls": "PHASE"}] , error happened with msg: Traceback (most recent call last): File "C:\Users\user\Desktop\PaddleOCR-2.6.0\ppocr\data\simple_dataset.py", line 138, in getitem outs = transform(data, self.ops) File "C:\Users\user\Desktop\PaddleOCR-2.6.0\ppocr\data\imaug__init.py", line 53, in transform data = op(data) File "C:\Users\user\Desktop\PaddleOCR-2.6.0\ppocr\data\imaug\label_ops.py", line 67, in call txt_tags = np.array(txt_tags, dtype=np.bool) File "C:\Users\user\anaconda3\envs\paddle310\lib\site-packages\numpy__init.py", line 324, in getattr raise AttributeError(former_attrs[attr]) AttributeError: module 'numpy' has no attribute 'bool'.
np.boolwas a deprecated alias for the builtinbool. To avoid this error in existing code, useboolby itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, usenp.bool_here. The aliases was originally deprecated in NumPy 1.20; for more details and guidance see the original release note at: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations. Did you mean: 'bool_'?[2024/03/13 19:48:44] ppocr ERROR: When parsing line C:\Users\user\Desktop\PaddleOCR-2.6.0\train_data\det\train\ENmotor35.jpg [{"transcription": "3-PHASE", "points": [[283, 77], [419, 77], [419, 99], [283, 99]], "difficult": false, "key_cls": "PHASE"}, {"transcription": "50HP", "points": [[175, 143], [242, 143], [242, 162], [175, 162]], "difficult": false, "key_cls": "HP"}, {"transcription": "37kW", "points": [[248, 141], [313, 141], [313, 161], [248, 161]], "difficult": false, "key_cls": "kW"}, {"transcription": "200LA", "points": [[426, 134], [511, 134], [511, 157], [426, 157]], "difficult": false, "key_cls": "FRAME"}, {"transcription": "POLES2", "points": [[94, 168], [224, 168], [224, 190], [94, 190]], "difficult": false, "key_cls": "POLE"}, {"transcription": "Hz50", "points": [[108, 187], [269, 187], [269, 212], [108, 212]], "difficult": false, "key_cls": "FREQUENCY"}, {"transcription": "380", "points": [[193, 211], [285, 211], [285, 233], [193, 233]], "difficult": false, "key_cls": "VOLTS"}, {"transcription": "70", "points": [[206, 237], [248, 237], [248, 263], [206, 263]], "difficult": false, "key_cls": "AMPS"}, {"transcription": "R.PM.2940", "points": [[80, 263], [264, 263], [264, 287], [80, 287]], "difficult": false, "key_cls": "RPM"}, {"transcription": "DATE 2011", "points": [[58, 375], [208, 375], [208, 405], [58, 405]], "difficult": false, "key_cls": "DATE"}] , error happened with msg: Traceback (most recent call last): File "C:\Users\user\Desktop\PaddleOCR-2.6.0\ppocr\data\simple_dataset.py", line 138, in getitem outs = transform(data, self.ops) File "C:\Users\user\Desktop\PaddleOCR-2.6.0\ppocr\data\imaug__init.py", line 53, in transform data = op(data) File "C:\Users\user\Desktop\PaddleOCR-2.6.0\ppocr\data\imaug\label_ops.py", line 67, in call txt_tags = np.array(txt_tags, dtype=np.bool) File "C:\Users\user\anaconda3\envs\paddle310\lib\site-packages\numpy__init.py", line 324, in getattr raise AttributeError(former_attrs[attr]) AttributeError: module 'numpy' has no attribute 'bool'.
np.boolwas a deprecated alias for the builtinbool. To avoid this error in existing code, useboolby itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, usenp.bool_here. The aliases was originally deprecated in NumPy 1.20; for more details and guidance see the original release note at: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations. Did you mean: 'bool_'?Exception in thread Thread-1 (_thread_loop): Traceback (most recent call last): File "C:\Users\user\Desktop\PaddleOCR-2.6.0\ppocr\data\simple_dataset.py", line 138, in getitem outs = transform(data, self.ops) File "C:\Users\user\Desktop\PaddleOCR-2.6.0\ppocr\data\imaug__init.py", line 53, in transform data = op(data) File "C:\Users\user\Desktop\PaddleOCR-2.6.0\ppocr\data\imaug\label_ops.py", line 67, in call txt_tags = np.array(txt_tags, dtype=np.bool) File "C:\Users\user\anaconda3\envs\paddle310\lib\site-packages\numpy__init.py", line 324, in getattr raise AttributeError(former_attrs[attr]) AttributeError: module 'numpy' has no attribute 'bool'.
np.boolwas a deprecated alias for the builtinbool. To avoid this error in existing code, useboolby itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, usenp.bool_here. The aliases was originally deprecated in NumPy 1.20; for more details and guidance see the original release note at: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations. Did you mean: 'bool_'?During handling of the above exception, another exception occurred:
Traceback (most recent call last): File "C:\Users\user\Desktop\PaddleOCR-2.6.0\ppocr\data\simple_dataset.py", line 153, in getitem return self.getitem(rnd_idx) File "C:\Users\user\Desktop\PaddleOCR-2.6.0\ppocr\data\simple_dataset.py", line 153, in getitem return self.getitem(rnd_idx) File "C:\Users\user\Desktop\PaddleOCR-2.6.0\ppocr\data\simple_dataset.py", line 153, in getitem return self.getitem(rnd_idx) [Previous line repeated 985 more times] File "C:\Users\user\Desktop\PaddleOCR-2.6.0\ppocr\data\simple_dataset.py", line 140, in getitem self.logger.error( File "C:\Users\user\anaconda3\envs\paddle310\lib\logging__init.py", line 1506, in error self._log(ERROR, msg, args, **kwargs) File "C:\Users\user\anaconda3\envs\paddle310\lib\logging__init__.py", line 1624, in _log self.handle(record) File "C:\Users\user\anaconda3\envs\paddle310\lib\logging__init.py", line 1634, in handle self.callHandlers(record) File "C:\Users\user\anaconda3\envs\paddle310\lib\logging\init.py", line 1696, in callHandlers hdlr.handle(record) File "C:\Users\user\anaconda3\envs\paddle310\lib\logging\init.py", line 968, in handle self.emit(record) File "C:\Users\user\anaconda3\envs\paddle310\lib\logging\init.py", line 1218, in emit StreamHandler.emit(self, record) File "C:\Users\user\anaconda3\envs\paddle310\lib\logging\init.py", line 1100, in emit msg = self.format(record) File "C:\Users\user\anaconda3\envs\paddle310\lib\logging\init.py", line 943, in format return fmt.format(record) File "C:\Users\user\anaconda3\envs\paddle310\lib\logging\init.py", line 679, in format if self.usesTime(): File "C:\Users\user\anaconda3\envs\paddle310\lib\logging\init__.py", line 647, in usesTime return self._style.usesTime() File "C:\Users\user\anaconda3\envs\paddle310\lib\logging\init__.py", line 424, in usesTime return self._fmt.find(self.asctime_search) >= 0 RecursionError: maximum recursion depth exceeded while calling a Python object 请尽量不要包含图片在问题中/Please try to not include the image in the issue.
Architecture: model_type: det algorithm: DB Transform: Backbone: name: MobileNetV3 scale: 0.5 model_name: large disable_se: True Neck: name: RSEFPN out_channels: 96 shortcut: True Head: name: DBHead k: 50
Loss: name: DBLoss balance_loss: true main_loss_type: DiceLoss alpha: 5 beta: 10 ohem_ratio: 3 Optimizer: name: Adam beta1: 0.9 beta2: 0.999 lr: name: Cosine learning_rate: 0.00005 warmup_epoch: 2 regularizer: name: L2 factor: 0 PostProcess: name: DBPostProcess thresh: 0.3 box_thresh: 0.6 max_candidates: 1000 unclip_ratio: 1.5 Metric: name: DetMetric main_indicator: hmean Train: dataset: name: SimpleDataSet data_dir: ./train_data/det label_file_list: