报错ValueError: zero-size array to reduction operation maximum which has no identity

[2257396011]

Description of the bug | 错误描述

2024-08-12 08:05:16.382 | ERROR | main:pdf_parse_main:135 - zero-size array to reduction operation maximum which has no identity
Traceback (most recent call last):

File "/home/founder/New/MinerU-master/demo/magic_pdf_parse_main.py", line 224, in
pdf_parse_main(pdf_path)
│ └ '/home/founder/New/pdf/1.pdf'
└ <function pdf_parse_main at 0x7fac4b7765f0>

File "/home/founder/New/MinerU-master/demo/magic_pdf_parse_main.py", line 117, in pdf_parse_main
pipe.pipe_analyze() # 解析
│ └ <function TXTPipe.pipe_analyze at 0x7fac4b776170>
└ <magic_pdf.pipe.TXTPipe.TXTPipe object at 0x7fac3ebcc880>

File "/home/founder/anaconda3/envs/MinerU/lib/python3.10/site-packages/magic_pdf/pipe/TXTPipe.py", line 20, in pipe_analyze
self.model_list = doc_analyze(self.pdf_bytes, ocr=False)
│ │ │ │ └ b'%PDF-1.7\r%\xe2\xe3\xcf\xd3\r\n308 0 obj\r<</Linearized 1/L 643506/O 310/E 111574/N 6/T 642787/H [ 660 557]>>\rendobj\r ...
│ │ │ └ <magic_pdf.pipe.TXTPipe.TXTPipe object at 0x7fac3ebcc880>
│ │ └ <function doc_analyze at 0x7fad6f68e9e0>
│ └ []
└ <magic_pdf.pipe.TXTPipe.TXTPipe object at 0x7fac3ebcc880>
File "/home/founder/anaconda3/envs/MinerU/lib/python3.10/site-packages/magic_pdf/model/doc_analyze_by_custom_model.py", line 119, in doc_analyze
result = custom_model(img)
│ └ array([[[255, 255, 255],
│ [255, 255, 255],
│ [255, 255, 255],
│ ...,
│ [255, 255, 255],
│ [255...
└ <magic_pdf.model.pdf_extract_kit.CustomPEKModel object at 0x7fac3ebcc730>
File "/home/founder/anaconda3/envs/MinerU/lib/python3.10/site-packages/magic_pdf/model/pdf_extract_kit.py", line 182, in call
for mf_img in dataloader:
└ <torch.utils.data.dataloader.DataLoader object at 0x7faab681c3a0>
File "/home/founder/.local/lib/python3.10/site-packages/torch/utils/data/dataloader.py", line 631, in next
data = self._next_data()
│ └ <function _SingleProcessDataLoaderIter._next_data at 0x7fabf422cc10>
└ <torch.utils.data.dataloader._SingleProcessDataLoaderIter object at 0x7faa0f90e560>
File "/home/founder/.local/lib/python3.10/site-packages/torch/utils/data/dataloader.py", line 675, in _next_data
data = self._dataset_fetcher.fetch(index) # may raise StopIteration
│ │ │ └ [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33...
│ │ └ <function _MapDatasetFetcher.fetch at 0x7fabf43b5d80>
│ └ <torch.utils.data._utils.fetch._MapDatasetFetcher object at 0x7faa0f90e260>
└ <torch.utils.data.dataloader._SingleProcessDataLoaderIter object at 0x7faa0f90e560>
File "/home/founder/.local/lib/python3.10/site-packages/torch/utils/data/_utils/fetch.py", line 51, in fetch
data = [self.dataset[idx] for idx in possibly_batched_index]
│ │ └ [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33...
│ └ <magic_pdf.model.pdf_extract_kit.MathDataset object at 0x7faab550fc70>
└ <torch.utils.data._utils.fetch._MapDatasetFetcher object at 0x7faa0f90e260>
File "/home/founder/.local/lib/python3.10/site-packages/torch/utils/data/_utils/fetch.py", line 51, in
data = [self.dataset[idx] for idx in possibly_batched_index]
│ │ │ └ 27
│ │ └ 27
│ └ <magic_pdf.model.pdf_extract_kit.MathDataset object at 0x7faab550fc70>
└ <torch.utils.data._utils.fetch._MapDatasetFetcher object at 0x7faa0f90e260>
File "/home/founder/anaconda3/envs/MinerU/lib/python3.10/site-packages/magic_pdf/model/pdf_extract_kit.py", line 82, in getitem
image = self.transform(raw_image)
│ │ └ <PIL.Image.Image image mode=RGB size=143x0 at 0x7FAAB6954460>
│ └ Compose(
│ <unimernet.processors.formula_processor.FormulaImageEvalProcessor object at 0x7faab550ff10>
│ )
└ <magic_pdf.model.pdf_extract_kit.MathDataset object at 0x7faab550fc70>
File "/home/founder/.local/lib/python3.10/site-packages/torchvision/transforms/transforms.py", line 95, in call
img = t(img)
│ └ <PIL.Image.Image image mode=RGB size=143x0 at 0x7FAAB6954460>
└ <unimernet.processors.formula_processor.FormulaImageEvalProcessor object at 0x7faab550ff10>
File "/home/founder/anaconda3/envs/MinerU/lib/python3.10/site-packages/unimernet/processors/formula_processor.py", line 164, in call
image = self.prepare_input(item)
│ │ └ <PIL.Image.Image image mode=RGB size=143x0 at 0x7FAAB6954460>
│ └ <function FormulaImageBaseProcessor.prepare_input at 0x7faadc049900>
└ <unimernet.processors.formula_processor.FormulaImageEvalProcessor object at 0x7faab550ff10>
File "/home/founder/anaconda3/envs/MinerU/lib/python3.10/site-packages/unimernet/processors/formula_processor.py", line 51, in prepare_input
img = self.crop_margin(img.convert("RGB"))
│ │ │ └ <function Image.convert at 0x7fac402824d0>
│ │ └ <PIL.Image.Image image mode=RGB size=143x0 at 0x7FAAB6954460>
│ └ <staticmethod(<function FormulaImageBaseProcessor.crop_margin at 0x7faadc04ab90>)>
└ <unimernet.processors.formula_processor.FormulaImageEvalProcessor object at 0x7faab550ff10>
File "/home/founder/anaconda3/envs/MinerU/lib/python3.10/site-packages/unimernet/processors/formula_processor.py", line 29, in crop_margin
max_val = data.max()
│ └ <method 'max' of 'numpy.ndarray' objects>
└ array([], shape=(0, 143), dtype=uint8)
File "/home/founder/.local/lib/python3.10/site-packages/numpy/core/_methods.py", line 41, in _amax
return umr_maximum(a, axis, None, out, keepdims, initial, where)
│ │ │ │ │ │ └ True
│ │ │ │ │ └
│ │ │ │ └ False
│ │ │ └ None
│ │ └ None
│ └ array([], shape=(0, 143), dtype=uint8)
└ <built-in method reduce of numpy.ufunc object at 0x7fad6e6cc440>

ValueError: zero-size array to reduction operation maximum which has no identity

How to reproduce the bug | 如何复现

magic_pdf_parse_main.py
软件版本0.7.x

Operating system | 操作系统

Linux

Python version | Python 版本

3.10

Software version | 软件版本 (magic-pdf --version)

0.6.x

Device mode | 设备模式

cuda

[myhloli]

可以上传一下出错的pdf文件？

[2257396011]

上传可能出现错误的pdf文件？

1.pdf
这个错有的时候会有，有的时候不会有，然后还有个问题就是这个pdf在版面分析的时候第二个表格总会识别成图片

[myhloli]

ValueError: zero-size array to reduction operation maximum which has no identity
这个报错未能复现，目前不知道触发条件，后面那个表格识别成图片的情况复现了，我们内部收集下排期修复。

[2257396011]

ValueError: zero-size array to reduction operation maximum which has no identity 这个报错未能复现，目前不知道触发条件，后面那个表格识别成图片的情况复现了，我们内部收集下排期修复。

好的，感觉可能是显存问题。

[2257396011]

ValueError: 零大小数组到缩减操作最大值没有身份 这个报错未能恢复现，目前不知道触发条件，后面那个表格识别成图片的情况恢复了，我们内部收集下排期修复。

你好，还有个问题也是针对这个pdf的，就是第四页的公式识别时间过长的问题，请问你们在复现的时候有这个问题么，我感觉可能是中文公式的问题。