PDF-Extract-Kit · Issues· 107 open

授权

`slanet-plus.onnx` has ONNX-prohibited nested-scope name collisions and fails to import in MIGraphX

[BUG] `table_parsing.py` 处理目录包含多张表格图片时 CUDA Out of Memory | CUDA OOM when parsing a directory with many table images

文档内容提取样例运行错误

Dna data

_pickle.UnpicklingError: invalid load key, 'v'. 错误

file not found

GPU进行OCR识别没有识别出内容，但是代码也没有报错信息， 而CPU能识别

PaddleOCR是2.7.x的版本

你好只用layout_detection的时候怎么获取相应的位置信息的json,目前只有一个 png图片输出。

Error of free(): double free detected in tcache2 was found when running with fastapi server

ImportError: generic_type: type "_gpuDeviceProperties" is already registered!

pdf转markdown报错

OpenMP Runtime Initialization Conflict (libiomp5md.dll) in scripts/ocr.py

ocr tasks issue(paddle-gpu&cuda&cudnn version compatibility issues))