运行表格识别DEMO报如下错误

[kevinwei1975]

https://github.com/opendatalab/PDF-Extract-Kit/blob/main/demo/TableRec/StructEqTable/Table_Recognition.ipynb 运行上述代码报错： 连续遇到三个问题， 第一个问题： from modelscope import snapshot_download model_dir = snapshot_download('wanderkid/PDF-Extract-Kit') 运行上述两行代码报错，说是缺少config.json这个文件，但是模型已经下载到本地了。 这个问题我没解决但是我发现，模型已经下载到本地了，我就用把路径添加到了代码即这行代码， model = build_model("C:/Users/kevin/.cache/modelscope/hub/wanderkid/PDF-Extract-Kit/models/TabRec/StructEqTable", max_new_tokens=4096, max_time=60) 他可以直接生成模型实例。 但是执行这行代码时报有病毒，提示如下： 木马：HEUR/QVM202.0.0DE9.Malware.Gen C:\Users\kevin\AppData\Local\Pandoc\pandoc.exe 如果忽略这个病毒，报如下错误， RuntimeError: Pandoc died with exitcode "65" during conversion: Error at "source" (line 1, column 90): unexpected end of input expecting \end{tabular} \begin{tabular}{|l|l|c|c|c|} \hline Label & Co-occurrence & Bonferroni & FDR & Color \ 即这行代码报错:markdown_code = convert_text(latex_code, 'md', format='latex')

[wangbinDL]

@sky-fly97 Please take a look at this issue.

[sky-fly97]

It's possible that you're running longer than the 60s maximum inference time you've set, and that's why the latex code is incomplete, so you can try increasing the max_time.