Closed amingze closed 1 year ago
解决了,paddlepaddle不要用最新的2.5.1的版本,可以用2.4.2的版本,并且降级paddlenlp的版本到2.5.0
pip uninstall paddlepaddle-gpu
python -m pip install paddlepaddle-gpu==2.4.1.post117 -f https://www.paddlepaddle.org.cn/whl/windows/mkl/avx/stable.html
pip uninstall paddlenlp
pip install paddlenlp==2.5.0 -i https://pypi.tuna.tsinghua.edu.cn/simple
以下是我的pip list
aiohttp 3.8.6
aiosignal 1.3.1
annotated-types 0.6.0
anyio 3.7.1
astor 0.8.1
async-timeout 4.0.3
attrdict 2.0.1
attrs 23.1.0
Babel 2.13.0
bce-python-sdk 0.8.90
beautifulsoup4 4.12.2
blinker 1.6.3
cachetools 5.3.1
certifi 2023.7.22
charset-normalizer 3.3.0
click 8.1.7
colorama 0.4.6
colorlog 6.7.0
contourpy 1.1.1
cssselect 1.2.0
cssutils 2.9.0
cycler 0.12.1
Cython 3.0.3
datasets 2.14.5
decorator 5.1.1
dill 0.3.4
et-xmlfile 1.1.0
exceptiongroup 1.1.3
fastapi 0.103.2
filelock 3.12.4
fire 0.5.0
Flask 3.0.0
flask-babel 4.0.0
fonttools 4.43.1
frozenlist 1.4.0
fsspec 2023.6.0
future 0.18.3
h11 0.14.0
httpcore 0.18.0
httpx 0.25.0
huggingface-hub 0.18.0
idna 3.4
imageio 2.31.5
imgaug 0.4.0
importlib-metadata 6.8.0
importlib-resources 6.1.0
itsdangerous 2.1.2
jieba 0.42.1
Jinja2 3.1.2
joblib 1.3.2
kiwisolver 1.4.5
lanms-neo 1.0.2
lazy_loader 0.3
lmdb 1.4.1
lxml 4.9.3
markdown-it-py 3.0.0
MarkupSafe 2.1.3
matplotlib 3.7.3
mdurl 0.1.2
multidict 6.0.4
multiprocess 0.70.12.2
networkx 3.1
numpy 1.24.4
onnx 1.14.1
opencv-contrib-python 4.6.0.66
opencv-python 4.6.0.66
openpyxl 3.1.2
opt-einsum 3.3.0
packaging 23.2
paddle-bfloat 0.1.7
paddle2onnx 1.0.6
paddlefsl 1.1.0
paddlenlp 2.5.0
paddleocr 2.7.0.3
paddlepaddle-gpu 2.4.2.post117
pandas 2.0.3
pdf2docx 0.5.6
Pillow 10.0.1
pip 23.2.1
Polygon3 3.0.9.1
premailer 3.10.0
protobuf 3.20.0
psutil 5.9.5
pyarrow 13.0.0
pyclipper 1.3.0.post5
pycryptodome 3.19.0
pydantic 2.4.2
pydantic_core 2.10.1
Pygments 2.16.1
PyMuPDF 1.20.2
pypandoc 1.11
pyparsing 3.1.1
python-dateutil 2.8.2
python-docx 1.0.1
pytz 2023.3.post1
PyWavelets 1.4.1
PyYAML 6.0.1
rapidfuzz 3.4.0
rarfile 4.1
requests 2.31.0
rich 13.6.0
safetensors 0.4.0
scikit-image 0.21.0
scikit-learn 1.3.1
scipy 1.10.1
sentencepiece 0.1.99
seqeval 1.2.2
setuptools 56.0.0
shapely 2.0.2
six 1.16.0
sniffio 1.3.0
soupsieve 2.5
starlette 0.27.0
tensorrt 8.6.1
termcolor 2.3.0
threadpoolctl 3.2.0
tifffile 2023.7.10
tqdm 4.66.1
typer 0.9.0
typing_extensions 4.8.0
tzdata 2023.3
urllib3 2.0.6
uvicorn 0.23.2
visualdl 2.5.3
Werkzeug 3.0.0
xxhash 3.4.1
yacs 0.1.8
yarl 1.9.2
zipp 3.17.0
以下是aistudio里的pip list
aiohttp 3.8.6
aiosignal 1.3.1
annotated-types 0.6.0
anyio 3.7.1
astor 0.8.1
async-timeout 4.0.3
attrdict 2.0.1
attrs 23.1.0
Babel 2.13.0
bce-python-sdk 0.8.90
beautifulsoup4 4.12.2
blinker 1.6.3
cachetools 5.3.1
certifi 2023.7.22
charset-normalizer 3.3.0
click 8.1.7
colorama 0.4.6
colorlog 6.7.0
contourpy 1.1.1
cssselect 1.2.0
cssutils 2.9.0
cycler 0.12.1
Cython 3.0.3
datasets 2.14.5
decorator 5.1.1
dill 0.3.4
et-xmlfile 1.1.0
exceptiongroup 1.1.3
fastapi 0.103.2
filelock 3.12.4
fire 0.5.0
Flask 3.0.0
flask-babel 4.0.0
fonttools 4.43.1
frozenlist 1.4.0
fsspec 2023.6.0
future 0.18.3
h11 0.14.0
httpcore 0.18.0
httpx 0.25.0
huggingface-hub 0.18.0
idna 3.4
imageio 2.31.5
imgaug 0.4.0
importlib-metadata 6.8.0
importlib-resources 6.1.0
itsdangerous 2.1.2
jieba 0.42.1
Jinja2 3.1.2
joblib 1.3.2
kiwisolver 1.4.5
lanms-neo 1.0.2
lazy_loader 0.3
lmdb 1.4.1
lxml 4.9.3
markdown-it-py 3.0.0
MarkupSafe 2.1.3
matplotlib 3.7.3
mdurl 0.1.2
multidict 6.0.4
multiprocess 0.70.12.2
networkx 3.1
numpy 1.24.4
onnx 1.14.1
opencv-contrib-python 4.6.0.66
opencv-python 4.6.0.66
openpyxl 3.1.2
opt-einsum 3.3.0
packaging 23.2
paddle-bfloat 0.1.7
paddle2onnx 1.0.6
paddlefsl 1.1.0
paddlenlp 2.5.0
paddleocr 2.7.0.3
paddlepaddle-gpu 2.4.2.post117
pandas 2.0.3
pdf2docx 0.5.6
Pillow 10.0.1
pip 23.2.1
Polygon3 3.0.9.1
premailer 3.10.0
protobuf 3.20.0
psutil 5.9.5
pyarrow 13.0.0
pyclipper 1.3.0.post5
pycryptodome 3.19.0
pydantic 2.4.2
pydantic_core 2.10.1
Pygments 2.16.1
PyMuPDF 1.20.2
pypandoc 1.11
pyparsing 3.1.1
python-dateutil 2.8.2
python-docx 1.0.1
pytz 2023.3.post1
PyWavelets 1.4.1
PyYAML 6.0.1
rapidfuzz 3.4.0
rarfile 4.1
requests 2.31.0
rich 13.6.0
safetensors 0.4.0
scikit-image 0.21.0
scikit-learn 1.3.1
scipy 1.10.1
sentencepiece 0.1.99
seqeval 1.2.2
setuptools 56.0.0
shapely 2.0.2
six 1.16.0
sniffio 1.3.0
soupsieve 2.5
starlette 0.27.0
tensorrt 8.6.1
termcolor 2.3.0
threadpoolctl 3.2.0
tifffile 2023.7.10
tqdm 4.66.1
typer 0.9.0
typing_extensions 4.8.0
tzdata 2023.3
urllib3 2.0.6
uvicorn 0.23.2
visualdl 2.5.3
Werkzeug 3.0.0
xxhash 3.4.1
yacs 0.1.8
yarl 1.9.2
zipp 3.17.0
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
[2023/10/13 09:41:54] ppocr INFO: ** ser config ** [2023/10/13 09:41:54] ppocr INFO: Architecture : [2023/10/13 09:41:54] ppocr INFO: Backbone : [2023/10/13 09:41:54] ppocr INFO: checkpoints : ./pretrained_model/ser_vi_layoutxlm_xfund_pretrained/best_accuracy [2023/10/13 09:41:54] ppocr INFO: mode : vi [2023/10/13 09:41:54] ppocr INFO: name : LayoutXLMForSer [2023/10/13 09:41:54] ppocr INFO: num_classes : 7 [2023/10/13 09:41:54] ppocr INFO: pretrained : True [2023/10/13 09:41:54] ppocr INFO: Transform : None [2023/10/13 09:41:54] ppocr INFO: algorithm : LayoutXLM [2023/10/13 09:41:54] ppocr INFO: model_type : kie [2023/10/13 09:41:54] ppocr INFO: Eval : [2023/10/13 09:41:54] ppocr INFO: dataset : [2023/10/13 09:41:54] ppocr INFO: data_dir : train_data/XFUND/zh_val/image [2023/10/13 09:41:54] ppocr INFO: label_file_list : ['train_data/XFUND/zh_val/val.json'] [2023/10/13 09:41:54] ppocr INFO: name : SimpleDataSet [2023/10/13 09:41:54] ppocr INFO: transforms : [2023/10/13 09:41:54] ppocr INFO: DecodeImage : [2023/10/13 09:41:54] ppocr INFO: channel_first : False [2023/10/13 09:41:54] ppocr INFO: img_mode : RGB [2023/10/13 09:41:54] ppocr INFO: VQATokenLabelEncode : [2023/10/13 09:41:54] ppocr INFO: algorithm : LayoutXLM [2023/10/13 09:41:54] ppocr INFO: class_path : train_data/XFUND/class_list_xfun.txt [2023/10/13 09:41:54] ppocr INFO: contains_re : False [2023/10/13 09:41:54] ppocr INFO: order_method : tb-yx [2023/10/13 09:41:54] ppocr INFO: use_textline_bbox_info : True [2023/10/13 09:41:54] ppocr INFO: VQATokenPad : [2023/10/13 09:41:54] ppocr INFO: max_seq_len : 512 [2023/10/13 09:41:54] ppocr INFO: return_attention_mask : True [2023/10/13 09:41:54] ppocr INFO: VQASerTokenChunk : [2023/10/13 09:41:54] ppocr INFO: max_seq_len : 512 [2023/10/13 09:41:54] ppocr INFO: Resize : [2023/10/13 09:41:54] ppocr INFO: size : [224, 224] [2023/10/13 09:41:54] ppocr INFO: NormalizeImage : [2023/10/13 09:41:54] ppocr INFO: mean : [123.675, 116.28, 103.53] [2023/10/13 09:41:54] ppocr INFO: order : hwc [2023/10/13 09:41:54] ppocr INFO: scale : 1 [2023/10/13 09:41:54] ppocr INFO: std : [58.395, 57.12, 57.375] [2023/10/13 09:41:54] ppocr INFO: ToCHWImage : None [2023/10/13 09:41:54] ppocr INFO: KeepKeys : [2023/10/13 09:41:54] ppocr INFO: keep_keys : ['input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'labels'] [2023/10/13 09:41:54] ppocr INFO: loader : [2023/10/13 09:41:54] ppocr INFO: batch_size_per_card : 8 [2023/10/13 09:41:54] ppocr INFO: drop_last : False [2023/10/13 09:41:54] ppocr INFO: num_workers : 4 [2023/10/13 09:41:54] ppocr INFO: shuffle : False [2023/10/13 09:41:54] ppocr INFO: Global : [2023/10/13 09:41:54] ppocr INFO: amp_custom_white_list : ['scale', 'concat', 'elementwise_add'] [2023/10/13 09:41:54] ppocr INFO: cal_metric_during_train : False [2023/10/13 09:41:54] ppocr INFO: d2s_train_image_shape : [3, 224, 224] [2023/10/13 09:41:54] ppocr INFO: epoch_num : 200 [2023/10/13 09:41:54] ppocr INFO: eval_batch_step : [0, 19] [2023/10/13 09:41:54] ppocr INFO: infer_img : ppstructure/docs/kie/input/zh_val_42.jpg [2023/10/13 09:41:54] ppocr INFO: kie_det_model_dir : None [2023/10/13 09:41:54] ppocr INFO: kie_rec_model_dir : None [2023/10/13 09:41:54] ppocr INFO: log_smooth_window : 10 [2023/10/13 09:41:54] ppocr INFO: print_batch_step : 10 [2023/10/13 09:41:54] ppocr INFO: save_epoch_step : 2000 [2023/10/13 09:41:54] ppocr INFO: save_inference_dir : None [2023/10/13 09:41:54] ppocr INFO: save_model_dir : ./output/ser_vi_layoutxlm_xfund_zh [2023/10/13 09:41:54] ppocr INFO: save_res_path : ./output/ser/xfund_zh/res [2023/10/13 09:41:54] ppocr INFO: seed : 2022 [2023/10/13 09:41:54] ppocr INFO: use_gpu : True [2023/10/13 09:41:54] ppocr INFO: use_visualdl : False [2023/10/13 09:41:54] ppocr INFO: Loss : [2023/10/13 09:41:54] ppocr INFO: key : backbone_out [2023/10/13 09:41:54] ppocr INFO: name : VQASerTokenLayoutLMLoss [2023/10/13 09:41:54] ppocr INFO: num_classes : 7 [2023/10/13 09:41:54] ppocr INFO: Metric : [2023/10/13 09:41:54] ppocr INFO: main_indicator : hmean [2023/10/13 09:41:54] ppocr INFO: name : VQASerTokenMetric [2023/10/13 09:41:54] ppocr INFO: Optimizer : [2023/10/13 09:41:54] ppocr INFO: beta1 : 0.9 [2023/10/13 09:41:54] ppocr INFO: beta2 : 0.999 [2023/10/13 09:41:54] ppocr INFO: lr : [2023/10/13 09:41:54] ppocr INFO: epochs : 200 [2023/10/13 09:41:54] ppocr INFO: learning_rate : 5e-05 [2023/10/13 09:41:54] ppocr INFO: name : Linear [2023/10/13 09:41:54] ppocr INFO: warmup_epoch : 2 [2023/10/13 09:41:54] ppocr INFO: name : AdamW [2023/10/13 09:41:54] ppocr INFO: regularizer : [2023/10/13 09:41:54] ppocr INFO: factor : 0.0 [2023/10/13 09:41:54] ppocr INFO: name : L2 [2023/10/13 09:41:54] ppocr INFO: PostProcess : [2023/10/13 09:41:54] ppocr INFO: class_path : train_data/XFUND/class_list_xfun.txt [2023/10/13 09:41:54] ppocr INFO: name : VQASerTokenLayoutLMPostProcess [2023/10/13 09:41:54] ppocr INFO: Train : [2023/10/13 09:41:54] ppocr INFO: dataset : [2023/10/13 09:41:54] ppocr INFO: data_dir : train_data/XFUND/zh_train/image [2023/10/13 09:41:54] ppocr INFO: label_file_list : ['train_data/XFUND/zh_train/train.json'] [2023/10/13 09:41:54] ppocr INFO: name : SimpleDataSet [2023/10/13 09:41:54] ppocr INFO: ratio_list : [1.0] [2023/10/13 09:41:54] ppocr INFO: transforms : [2023/10/13 09:41:54] ppocr INFO: DecodeImage : [2023/10/13 09:41:54] ppocr INFO: channel_first : False [2023/10/13 09:41:54] ppocr INFO: img_mode : RGB [2023/10/13 09:41:54] ppocr INFO: VQATokenLabelEncode : [2023/10/13 09:41:54] ppocr INFO: algorithm : LayoutXLM [2023/10/13 09:41:54] ppocr INFO: class_path : train_data/XFUND/class_list_xfun.txt [2023/10/13 09:41:54] ppocr INFO: contains_re : False [2023/10/13 09:41:54] ppocr INFO: order_method : tb-yx [2023/10/13 09:41:54] ppocr INFO: use_textline_bbox_info : True [2023/10/13 09:41:54] ppocr INFO: VQATokenPad : [2023/10/13 09:41:54] ppocr INFO: max_seq_len : 512 [2023/10/13 09:41:54] ppocr INFO: return_attention_mask : True [2023/10/13 09:41:54] ppocr INFO: VQASerTokenChunk : [2023/10/13 09:41:54] ppocr INFO: max_seq_len : 512 [2023/10/13 09:41:54] ppocr INFO: Resize : [2023/10/13 09:41:54] ppocr INFO: size : [224, 224] [2023/10/13 09:41:54] ppocr INFO: NormalizeImage : [2023/10/13 09:41:54] ppocr INFO: mean : [123.675, 116.28, 103.53] [2023/10/13 09:41:54] ppocr INFO: order : hwc [2023/10/13 09:41:54] ppocr INFO: scale : 1 [2023/10/13 09:41:54] ppocr INFO: std : [58.395, 57.12, 57.375] [2023/10/13 09:41:54] ppocr INFO: ToCHWImage : None [2023/10/13 09:41:54] ppocr INFO: KeepKeys : [2023/10/13 09:41:54] ppocr INFO: keep_keys : ['input_ids', 'bbox', 'attention_mask', 'token_type_ids', 'image', 'labels'] [2023/10/13 09:41:54] ppocr INFO: loader : [2023/10/13 09:41:54] ppocr INFO: batch_size_per_card : 8 [2023/10/13 09:41:54] ppocr INFO: drop_last : False [2023/10/13 09:41:54] ppocr INFO: num_workers : 4 [2023/10/13 09:41:54] ppocr INFO: shuffle : True [2023/10/13 09:41:54] ppocr INFO: train with paddle 2.5.1 and device Place(gpu:0) [2023-10-13 09:41:55,592] [ INFO] - Loading configuration file ./pretrained_model/ser_vi_layoutxlm_xfund_pretrained/best_accuracy\config.json [2023-10-13 09:41:55,593] [ INFO] - Loading weights file ./pretrained_model/ser_vi_layoutxlm_xfund_pretrained/best_accuracy\model_state.pdparams [2023-10-13 09:41:56,150] [ INFO] - Loaded weights file from disk, setting weights to model. W1013 09:41:56.152549 20880 gpu_resources.cc:119] Please NOTE: device: 0, GPU Compute Capability: 8.9, Driver API Version: 12.2, Runtime API Version: 11.8 W1013 09:41:56.156495 20880 gpu_resources.cc:149] device: 0, cuDNN Version: 8.9. [2023-10-13 09:41:58,145] [ INFO] - All model checkpoint weights were used when initializing LayoutXLMForTokenClassification.
[2023-10-13 09:41:58,145] [ INFO] - All the weights of LayoutXLMForTokenClassification were initialized from the model checkpoint at ./pretrained_model/ser_vi_layoutxlm_xfund_pretrained/best_accuracy. If your task is similar to the task the model of the checkpoint was trained on, you can already use LayoutXLMForTokenClassification for predictions without further training. [2023/10/13 09:41:58] ppocr INFO: resume from ./pretrained_model/ser_vi_layoutxlm_xfund_pretrained/best_accuracy [2023-10-13 09:42:01,419] [ INFO] - Already cached C:\Users\amingze.paddlenlp\models\layoutxlm-base-uncased\sentencepiece.bpe.model [2023-10-13 09:42:01,703] [ INFO] - tokenizer config file saved in C:\Users\amingze.paddlenlp\models\layoutxlm-base-uncased\tokenizer_config.json [2023-10-13 09:42:01,704] [ INFO] - Special tokens file saved in C:\Users\amingze.paddlenlp\models\layoutxlm-base-uncased\special_tokens_map.json [2023-10-13 09:42:01,706] [ INFO] - Loading configuration file ./pretrained_model/re_vi_layoutxlm_xfund_pretrained/best_accuracy\config.json [2023-10-13 09:42:01,707] [ INFO] - Loading weights file ./pretrained_model/re_vi_layoutxlm_xfund_pretrained/best_accuracy\model_state.pdparams [2023-10-13 09:42:02,401] [ INFO] - Loaded weights file from disk, setting weights to model. [2023-10-13 09:42:03,304] [ INFO] - All model checkpoint weights were used when initializing LayoutXLMForRelationExtraction.
[2023-10-13 09:42:03,304] [ INFO] - All the weights of LayoutXLMForRelationExtraction were initialized from the model checkpoint at ./pretrained_model/re_vi_layoutxlm_xfund_pretrained/best_accuracy. If your task is similar to the task the model of the checkpoint was trained on, you can already use LayoutXLMForRelationExtraction for predictions without further training. [2023/10/13 09:42:03] ppocr INFO: resume from ./pretrained_model/re_vi_layoutxlm_xfund_pretrained/best_accuracy Corrupt JPEG data: premature end of data segment Traceback (most recent call last): File "D:\code\paddle\PaddleOCR\tools\infer_kie_token_ser_re.py", line 217, in
result = ser_re_engine(data)
^^^^^^^^^^^^^^^^^^^
File "D:\code\paddle\PaddleOCR\tools\infer_kie_token_ser_re.py", line 151, in call
preds = self.model(re_input)
^^^^^^^^^^^^^^^^^^^^
File "C:\Users\amingze\anaconda3\Lib\site-packages\paddle\nn\layer\layers.py", line 1254, in call
return self.forward(*inputs, kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\code\paddle\PaddleOCR\ppocr\modeling\architectures\base_model.py", line 86, in forward
x = self.backbone(x)
^^^^^^^^^^^^^^^^
File "C:\Users\amingze\anaconda3\Lib\site-packages\paddle\nn\layer\layers.py", line 1254, in call
return self.forward(*inputs, *kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\code\paddle\PaddleOCR\ppocr\modeling\backbones\vqa_layoutlm.py", line 227, in forward
x = self.model(
^^^^^^^^^^^
File "C:\Users\amingze\anaconda3\Lib\site-packages\paddle\nn\layer\layers.py", line 1254, in call
return self.forward(inputs, kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\amingze\anaconda3\Lib\site-packages\paddlenlp\transformers\layoutxlm\modeling.py", line 1330, in forward
loss, pred_relations = self.extractor(sequence_output, entities, relations)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\amingze\anaconda3\Lib\site-packages\paddle\nn\layer\layers.py", line 1254, in call
return self.forward(*inputs, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\amingze\anaconda3\Lib\site-packages\paddlenlp\transformers\layoutxlm\modeling.py", line 1224, in forward
relations, entities = self.build_relation(relations, entities)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\amingze\anaconda3\Lib\site-packages\paddlenlp\transformers\layoutxlm\modeling.py", line 1186, in build_relation
negative_relations = all_possible_relations[negative_mask]