ValueError: (InvalidArgument) x dim number should greater than 0, but received value is: 0 [Hint: Expected x_dim > 0, but received x_dim:0 <= 0:0.] #11261

danghoangnhan · 2023-11-15T18:56:13Z

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

系统环境/System Environment：
- ubuntu 22.04
- cuda:11.8
版本号/Version:2.7：Paddle： PaddleOCR：
问题相关组件/Related components：
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==2.13.1
bce-python-sdk==0.8.96
blinker==1.7.0
cachetools==5.3.2
certifi==2023.7.22
charset-normalizer==3.3.2
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==3.0.5
datasets==2.14.7
decorator==5.1.1
dill==0.3.4
et-xmlfile==1.1.0
exceptiongroup==1.1.3
fastapi==0.104.1
filelock==3.13.1
flask==3.0.0
Flask-Babel==2.0.0
fonttools==4.44.0
frozenlist==1.4.0
fsspec==2023.10.0
future==0.18.3
h11==0.14.0
httpcore==1.0.2
httpx==0.25.1
huggingface-hub==0.19.3
idna==3.4
imageio==2.32.0
imgaug==0.4.0
importlib-metadata==6.8.0
importlib-resources==6.1.1
itsdangerous==2.1.2
jieba==0.42.1
Jinja==23.1.2
joblib==1.3.2
kiwisolver==1.4.5
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==1.15.0
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
paddle2onnx==1.1.0
paddlefsl==1.1.0
paddlenlp==2.5.2
paddlepaddle-gpu==2.5.2
pandas==2.0.3
Pillow==10.1.0
pip==20.0.2
pkg-resources==0.0.0
premailer==3.10.0
protobuf==3.20.2
psutil==5.9.6
pyarrow==14.0.1
pyarrow-hotfix==0.5
pyclipper==1.3.0.post5
pycryptodome==3.19.0
pydantic==2.5.1
pydantic-core==2.14.3
Pygments==2.16.1
PyMuPDF==1.20.2
pypandoc==1.12
pyparsing==3.1.1
python-dateutil==2.8.2
python-docx==1.1.0
pytz==2023.3.post1
PyWavelets==1.4.1
PyYAML==6.0.1
rapidfuzz==3.5.2
rarfile==4.1
requests==2.31.0
rich==13.7.0
safetensors==0.4.0
scikit-image==0.21.0
scikit-learn==1.3.2
scipy==1.10.1
sentencepiece==0.1.99
seqeval==1.2.2
setuptools==44.0.0
shapely==2.0.2
six==1.16.0
sniffio==1.3.0
starlette==0.27.0
threadpoolctl==3.2.0
tifffile==2023.7.10
tqdm==4.66.1
typer==0.9.0
typing-extensions==4.8.0
tzdata==2023.3
urllib==32.1.0
uvicorn==0.24.0.post1
visualdl==2.5.3
werkzeug==3.0.1
wheel==0.34.2
xxhash==3.4.1
yacs==0.1.8
yarl==1.9.2
zipp==3.17.0
运行指令/Command Code：

  python3 ./tools/infer_kie_token_ser_re.py \
  -c configs/kie/vi_layoutxlm/re_vi_layoutxlm_xfund_zh.yml \
  -o Architecture.Backbone.checkpoints=./output/re_vi_layoutxlm_xfund_zh/best_accuracy/ \
  Global.infer_img=./train_data/XFUND/zh_val/image/ \
  -c_ser configs/kie/vi_layoutxlm/ser_vi_layoutxlm_xfund_zh.yml \
  -o_ser Architecture.Backbone.checkpoints=output/re_vi_layoutxlm_xfund_zh/best_accuracy/

完整报错/Complete Error Message：

[2023/11/16 02:39:50] ppocr INFO: resume from ./output/re_vi_layoutxlm_xfund_zh/best_accuracy/
Traceback (most recent call last):
  File "./tools/infer_kie_token_ser_re.py", line 217, in <module>
    result = ser_re_engine(data)
  File "./tools/infer_kie_token_ser_re.py", line 151, in __call__
    preds = self.model(re_input)
  File "/home/daniel/PaddleOCR/venv/lib/python3.8/site-packages/paddle/nn/layer/layers.py", line 1254, in __call__
    return self.forward(*inputs, **kwargs)
  File "/home/daniel/PaddleOCR/ppocr/modeling/architectures/base_model.py", line 86, in forward
    x = self.backbone(x)
  File "/home/daniel/PaddleOCR/venv/lib/python3.8/site-packages/paddle/nn/layer/layers.py", line 1254, in __call__
    return self.forward(*inputs, **kwargs)
  File "/home/daniel/PaddleOCR/ppocr/modeling/backbones/vqa_layoutlm.py", line 227, in forward
    x = self.model(
  File "/home/daniel/PaddleOCR/venv/lib/python3.8/site-packages/paddle/nn/layer/layers.py", line 1254, in __call__
    return self.forward(*inputs, **kwargs)
  File "/home/daniel/PaddleOCR/venv/lib/python3.8/site-packages/paddlenlp/transformers/layoutxlm/modeling.py", line 1412, in forward
    loss, pred_relations = self.extractor(sequence_output, entities, relations)
  File "/home/daniel/PaddleOCR/venv/lib/python3.8/site-packages/paddle/nn/layer/layers.py", line 1254, in __call__
    return self.forward(*inputs, **kwargs)
  File "/home/daniel/PaddleOCR/venv/lib/python3.8/site-packages/paddlenlp/transformers/layoutxlm/modeling.py", line 1304, in forward
    relations, entities = self.build_relation(relations, entities)
  File "/home/daniel/PaddleOCR/venv/lib/python3.8/site-packages/paddlenlp/transformers/layoutxlm/modeling.py", line 1248, in build_relation
    all_possible_relations = paddle.stack(
  File "/home/daniel/PaddleOCR/venv/lib/python3.8/site-packages/paddle/tensor/manipulation.py", line 1842, in stack
    return _C_ops.stack(x, axis)
ValueError: (InvalidArgument) x dim number should greater than 0, but received value is: 0
  [Hint: Expected x_dim > 0, but received x_dim:0 <= 0:0.] (at ../paddle/phi/backends/gpu/gpu_launch_config.h:180)

The text was updated successfully, but these errors were encountered:

danghoangnhan · 2023-11-17T08:17:07Z

i followed the RE instruction in kie document and got this problem

gr8nishan · 2023-11-22T10:57:21Z

I am also facing the same issue in some documents...

heguyu · 2023-11-27T02:52:57Z

also

c-avan · 2023-12-12T05:08:54Z

@danghoangnhan 问题解决了吗，我也遇到这样的类似错了
ValueError: (InvalidArgument) x dim number should greater than 0, but received value is: 0
[Hint: Expected x_dim > 0, but received x_dim:0 <= 0:0.] (at ../paddle/phi/backends/gpu/gpu_launch_config.h:175)

Joheun-Kang · 2023-12-15T00:51:59Z

Got the same error, do we have any update on this issue?

XiYuan68 · 2024-02-02T05:29:49Z

same error when training RE models

XiYuan68 · 2024-02-21T06:28:30Z

same error when training RE models

PaddleOCR/ppocr/data/imaug/vqa/token/vqa_token_relation.py

Lines 35 to 49 in 0525f6b

    
           kv_relations = [] 
        
           for rel in relations: 
        
               pair = [id2label[rel[0]], id2label[rel[1]]] 
        
               if pair == ["question", "answer"]: 
        
                   kv_relations.append({ 
        
                       "head": entity_id_to_index_map[rel[0]], 
        
                       "tail": entity_id_to_index_map[rel[1]] 
        
                   }) 
        
               elif pair == ["answer", "question"]: 
        
                   kv_relations.append({ 
        
                       "head": entity_id_to_index_map[rel[1]], 
        
                       "tail": entity_id_to_index_map[rel[0]] 
        
                   }) 
        
               else: 
        
                   continue

Bug found, when building relations in dataloader, labels of entity are limited to "question" and "answer", custom dataset with other labels will be ignored and no relations built.

I modified it into:

        kv_relations = []
        for rel in relations:
            # assume labeled relations are always in the format of head-tail
            try:
                kv_relations.append({
                    "head": entity_id_to_index_map[rel[0]],
                    "tail": entity_id_to_index_map[rel[1]]
                })
            except Exception:
                pass

            # pair = [id2label[rel[0]], id2label[rel[1]]]
            # if pair == ["question", "answer"]:
            #     kv_relations.append({
            #         "head": entity_id_to_index_map[rel[0]],
            #         "tail": entity_id_to_index_map[rel[1]]
            #     })
            # elif pair == ["answer", "question"]:
            #     kv_relations.append({
            #         "head": entity_id_to_index_map[rel[1]],
            #         "tail": entity_id_to_index_map[rel[0]]
            #     })
            # else:
            #     continue
        # print('kv_relations:', kv_relations)

there is a try-except because sometimes entity_id_to_index_map[rel[1]] raises a keyerror

And still there will be (very) few invalid batches, I add a try-except in this for loop, the model starts training:

PaddleOCR/tools/program.py

Line 269 in 0525f6b

for idx, batch in enumerate(train_dataloader):

BB86818BB · 2024-03-08T11:37:00Z

同样的问题，训练的时候修改vqa_token_relation.py后可以了，但是预测的时候依然报错，完整的错误信息如下：

Traceback (most recent call last):
File "/home/ocr/PaddleOCR-release-2.7.1/./tools/infer_kie_token_ser_re.py", line 218, in
result = ser_re_engine(data)
File "/home/ocr/PaddleOCR-release-2.7.1/./tools/infer_kie_token_ser_re.py", line 152, in call
preds = self.model(re_input)
File "/usr/local/lib/python3.10/site-packages/paddle/nn/layer/layers.py", line 1429, in call
return self.forward(*inputs, **kwargs)
File "/home/ocr/PaddleOCR-release-2.7.1/ppocr/modeling/architectures/base_model.py", line 86, in forward
x = self.backbone(x)
File "/usr/local/lib/python3.10/site-packages/paddle/nn/layer/layers.py", line 1429, in call
return self.forward(*inputs, **kwargs)
File "/home/ocr/PaddleOCR-release-2.7.1/ppocr/modeling/backbones/vqa_layoutlm.py", line 220, in forward
x = self.model(
File "/usr/local/lib/python3.10/site-packages/paddle/nn/layer/layers.py", line 1429, in call
return self.forward(*inputs, **kwargs)
File "/usr/local/lib/python3.10/site-packages/paddlenlp/transformers/layoutxlm/modeling.py", line 1329, in forward
loss, pred_relations = self.extractor(sequence_output, entities, relations)
File "/usr/local/lib/python3.10/site-packages/paddle/nn/layer/layers.py", line 1429, in call
return self.forward(*inputs, **kwargs)
File "/usr/local/lib/python3.10/site-packages/paddlenlp/transformers/layoutxlm/modeling.py", line 1223, in forward
relations, entities = self.build_relation(relations, entities)
File "/usr/local/lib/python3.10/site-packages/paddlenlp/transformers/layoutxlm/modeling.py", line 1167, in build_relation
all_possible_relations = paddle.stack(
File "/usr/local/lib/python3.10/site-packages/paddle/tensor/manipulation.py", line 2001, in stack
return _C_ops.stack(x, axis)
ValueError: (InvalidArgument) x dim number should greater than 0, but received value is: 0
[Hint: Expected x_dim > 0, but received x_dim:0 <= 0:0.] (at ../paddle/phi/backends/gpu/gpu_launch_config.h:166)

mxhyxym · 2024-03-09T07:24:09Z

训练re模型的时候我也遇到这个问题，最后解决是通过修改数据集的label的顺序解决的。我是将question排在answer前面，就是对label排个序。我猜测是要保证linking=[[a,b]]的a需要放在b的前面。最后数据集长这样：[{id:1, linking:[[1, 2]]},{id:2, linking:[[1, 2]]}]，而不能[{id:2, linking:[[1, 2]]}， {id:1, linking:[[1, 2]]}]

Mehdi-Ben-Souda · 2024-05-30T10:04:41Z

Is there any news , i am still struggling with this issue in the prediction .

Royhuiy · 2024-06-20T05:00:46Z

训练re模型的时候我也遇到这个问题，最后解决是通过修改数据集的label的顺序解决的。我是将question排在answer前面，就是对label排个序。我猜测是要保证linking=[[a,b]]的a需要放在b的前面。最后数据集长这样：[{id:1, linking:[[1, 2]]},{id:2, linking:[[1, 2]]}]，而不能[{id:2, linking:[[1, 2]]}， {id:1, linking:[[1, 2]]}]

你好，请问方便提供带linking的1-2条数据集，我想参考一下

Royhuiy · 2024-06-20T05:08:22Z

@danghoangnhan 问题解决了吗，我也遇到这样的类似错了 ValueError: (InvalidArgument) x dim number should greater than 0, but received value is: 0 [Hint: Expected x_dim > 0, but received x_dim:0 <= 0:0.] (at ../paddle/phi/backends/gpu/gpu_launch_config.h:175)

你好，请问你解决这个问题了吗？

Royhuiy · 2024-06-20T05:17:51Z

i followed the RE instruction in kie document and got this problem

same error, how did you fix this?

freezehe · 2024-08-05T02:24:59Z

我也遇到了这个问题，请问怎么解决的？

danghoangnhan · 2024-08-06T08:07:36Z

Hi everyone, for this problem, I used several tricks to fix the IDs:

The IDs must start from 1.
The answer ID must be greater than the question ID.
In the linking part, the question must be on the left side, for example, linking: [[question_id, answer_id]].

Hope it helps!

turgutguvercin · 2024-12-26T06:55:30Z

If you're using a custom recognition model and have updated the dictionary, the model may still use the default dictionary. To apply your changes, you need to update the dictionary directory manually.

python3 predict_kie_token_ser_re.py --kie_algorithm=LayoutXLM
--re_model_dir=../../inference/re_vi_layoutxlm_xfund_infer --ser_model_dir=../../inference/ser_vi_layoutxlm_xfund_infer
--use_visual_backbone=False --image_dir=../../iski_125.jpg
--ser_dict_path=../../train_data/XFUND/class_list_xfun.txt --vis_font_path=../../fonts/Arial.ttf
--det_model_dir=../../inference_model_det_ppocr_v4_custom--rec_char_dict_path=../../dict/custom_dict.txt
--rec_model_dir=../../inference_model_rec_ppocr_v3_custom`
--ocr_order_method="tb-yx"

--rec_char_dict_path=../../dict/custom_dict.txt this line will not work that's why the error happens change dict path manually.

paddle-bot bot assigned tink2123 Feb 26, 2024

GreatV added the bug Something isn't working label Jun 20, 2024

freezehe mentioned this issue Aug 5, 2024

kie预测ser+re的时候报这个错误ValueError: (InvalidArgument) x dim number should greater than 0, but received value is: 0 #13589

Closed

4 tasks

chairc mentioned this issue Nov 6, 2024

KIE: re train error #9184

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

ValueError: (InvalidArgument) x dim number should greater than 0, but received value is: 0 [Hint: Expected x_dim > 0, but received x_dim:0 <= 0:0.] #11261

ValueError: (InvalidArgument) x dim number should greater than 0, but received value is: 0 [Hint: Expected x_dim > 0, but received x_dim:0 <= 0:0.] #11261

danghoangnhan commented Nov 15, 2023 •

edited

Loading

danghoangnhan commented Nov 17, 2023 •

edited

Loading

gr8nishan commented Nov 22, 2023

heguyu commented Nov 27, 2023

c-avan commented Dec 12, 2023

Joheun-Kang commented Dec 15, 2023

XiYuan68 commented Feb 2, 2024

XiYuan68 commented Feb 21, 2024 •

edited

Loading

BB86818BB commented Mar 8, 2024

mxhyxym commented Mar 9, 2024

Mehdi-Ben-Souda commented May 30, 2024

Royhuiy commented Jun 20, 2024

Royhuiy commented Jun 20, 2024

Royhuiy commented Jun 20, 2024

freezehe commented Aug 5, 2024

danghoangnhan commented Aug 6, 2024

turgutguvercin commented Dec 26, 2024

ValueError: (InvalidArgument) x dim number should greater than 0, but received value is: 0 [Hint: Expected x_dim > 0, but received x_dim:0 <= 0:0.] #11261

ValueError: (InvalidArgument) x dim number should greater than 0, but received value is: 0 [Hint: Expected x_dim > 0, but received x_dim:0 <= 0:0.] #11261

Comments

danghoangnhan commented Nov 15, 2023 • edited Loading

danghoangnhan commented Nov 17, 2023 • edited Loading

gr8nishan commented Nov 22, 2023

heguyu commented Nov 27, 2023

c-avan commented Dec 12, 2023

Joheun-Kang commented Dec 15, 2023

XiYuan68 commented Feb 2, 2024

XiYuan68 commented Feb 21, 2024 • edited Loading

BB86818BB commented Mar 8, 2024

mxhyxym commented Mar 9, 2024

Mehdi-Ben-Souda commented May 30, 2024

Royhuiy commented Jun 20, 2024

Royhuiy commented Jun 20, 2024

Royhuiy commented Jun 20, 2024

freezehe commented Aug 5, 2024

danghoangnhan commented Aug 6, 2024

turgutguvercin commented Dec 26, 2024

danghoangnhan commented Nov 15, 2023 •

edited

Loading

danghoangnhan commented Nov 17, 2023 •

edited

Loading

XiYuan68 commented Feb 21, 2024 •

edited

Loading