magic-pdf pdf-command --pdf "pdf_path" --inside_model true -> Segmentation fault (core dumped)

[Invariant0502]

Description of the bug | 错误描述

按流程复现 运行magic-pdf pdf-command --pdf "pdf_path" --inside_model true 报错如下：

C++ Traceback (most recent call last):

0 at::_ops::conv2d::call(at::Tensor const&, at::Tensor const&, std::optionalat::Tensor const&, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, c10::SymInt) 1 at::native::conv2d_symint(at::Tensor const&, at::Tensor const&, std::optionalat::Tensor const&, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, c10::SymInt) 2 at::_ops::convolution::call(at::Tensor const&, at::Tensor const&, std::optionalat::Tensor const&, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, bool, c10::ArrayRefc10::SymInt, c10::SymInt) 3 at::_ops::convolution::redispatch(c10::DispatchKeySet, at::Tensor const&, at::Tensor const&, std::optionalat::Tensor const&, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, bool, c10::ArrayRefc10::SymInt, c10::SymInt) 4 at::native::convolution(at::Tensor const&, at::Tensor const&, std::optionalat::Tensor const&, c10::ArrayRef, c10::ArrayRef, c10::ArrayRef, bool, c10::ArrayRef, long) 5 at::_ops::_convolution::call(at::Tensor const&, at::Tensor const&, std::optionalat::Tensor const&, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, bool, c10::ArrayRefc10::SymInt, c10::SymInt, bool, bool, bool, bool) 6 at::native::_convolution(at::Tensor const&, at::Tensor const&, std::optionalat::Tensor const&, c10::ArrayRef, c10::ArrayRef, c10::ArrayRef, bool, c10::ArrayRef, long, bool, bool, bool, bool) 7 at::_ops::cudnn_convolution::call(at::Tensor const&, at::Tensor const&, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, c10::SymInt, bool, bool, bool) 8 at::native::cudnn_convolution(at::Tensor const&, at::Tensor const&, c10::ArrayRef, c10::ArrayRef, c10::ArrayRef, long, bool, bool, bool)

---

Error Message Summary:

FatalError: Segmentation fault is detected by the operating system. [TimeInfo: *** Aborted at 1722363122 (unix time) try "date -d @1722363122" if you are using GNU date ***] [SignalInfo: *** SIGSEGV (@0x20000002ef4) received by PID 991495 (TID 0x7ffaa1cc54c0) from PID 12020 ***]

Segmentation fault (core dumped)

How to reproduce the bug | 如何复现

按流程

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

[drunkpig]

@Invariant0502 "pdf_path" in command line is a real path to a pdf file in your disk. For example :

magicpdf --pdf  /mnt/data/mybook.pdf  --inside_model true

[Invariant0502]

@Invariant0502 "pdf_path" in command line is a real path to a pdf file in your disk. For example :

magicpdf --pdf  /mnt/data/mybook.pdf  --inside_model true

Of course, I did this, and it reported this error after running for a period of time.

[xzxjysgd]

我也是这样的：等待回复！！ [08/02 10:56:57 d2.checkpoint.detection_checkpoint]: [DetectionCheckpointer] Loading from /home/yunke2024/MinerU/PDF-Extract-Kit/models/Layout/model_final.pth ... [08/02 10:56:57 fvcore.common.checkpoint]: [Checkpointer] Loading from /home/yunke2024/MinerU/PDF-Extract-Kit/models/Layout/model_final.pth ... 2024-08-02 10:56:58.235 | INFO | magic_pdf.model.pdf_extract_kit:init:132 - DocAnalysis init done! 2024-08-02 10:56:58.235 | INFO | magic_pdf.model.doc_analyze_by_custom_model:custom_model_init:92 - model init cost: 33.7154860496521

---

C++ Traceback (most recent call last):

0 at::_ops::conv2d::call(at::Tensor const&, at::Tensor const&, std::optionalat::Tensor const&, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, c10::SymInt) 1 at::native::conv2d_symint(at::Tensor const&, at::Tensor const&, std::optionalat::Tensor const&, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, c10::SymInt) 2 at::_ops::convolution::call(at::Tensor const&, at::Tensor const&, std::optionalat::Tensor const&, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, bool, c10::ArrayRefc10::SymInt, c10::SymInt) 3 at::_ops::convolution::redispatch(c10::DispatchKeySet, at::Tensor const&, at::Tensor const&, std::optionalat::Tensor const&, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, bool, c10::ArrayRefc10::SymInt, c10::SymInt) 4 at::native::convolution(at::Tensor const&, at::Tensor const&, std::optionalat::Tensor const&, c10::ArrayRef, c10::ArrayRef, c10::ArrayRef, bool, c10::ArrayRef, long) 5 at::_ops::_convolution::call(at::Tensor const&, at::Tensor const&, std::optionalat::Tensor const&, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, bool, c10::ArrayRefc10::SymInt, c10::SymInt, bool, bool, bool, bool) 6 at::native::_convolution(at::Tensor const&, at::Tensor const&, std::optionalat::Tensor const&, c10::ArrayRef, c10::ArrayRef, c10::ArrayRef, bool, c10::ArrayRef, long, bool, bool, bool, bool) 7 at::_ops::cudnn_convolution::call(at::Tensor const&, at::Tensor const&, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, c10::ArrayRefc10::SymInt, c10::SymInt, bool, bool, bool) 8 at::native::cudnn_convolution(at::Tensor const&, at::Tensor const&, c10::ArrayRef, c10::ArrayRef, c10::ArrayRef, long, bool, bool, bool)

---

Error Message Summary:

FatalError: Segmentation fault is detected by the operating system. [TimeInfo: *** Aborted at 1722567420 (unix time) try "date -d @1722567420" if you are using GNU date ***] [SignalInfo: *** SIGSEGV (@0x20000002ef4) received by PID 1026855 (TID 0x7f1d7a435740) from PID 12020 ***]

Segmentation fault (core dumped)

[mynewstart]

我也是

[Invariant0502]

The issue was resolved after updating to version 6.1.2b1.

[mynewstart]

https://github.com/opendatalab/MinerU/releases/tag/magic_pdf-0.6.2b1-released is this version?

[mynewstart]

Solved this by updating toch2.3.1+cu118 to torch2.3.1+cu121

[GuokaiLiu]

Solved this by updating toch2.3.1+cu118 to torch2.3.1+cu121

How to install torch2.3.1+cu121?