YeYuheng
YeYuheng
Yes, I encountered the same problem. After reading the source code, I found that I could only use the PDF on the cloud. like a shit.......
> I'm also facing the same issue as well. I've tried Python 3.11 and Python 3.12, no difference. Using version 1.6. hello ，what is mean 1.6？ python 3.16？
> RAPTOR 默认配置为使用该`collapse_tree`方法，因为它通常能提供更好的性能。如果要禁用该`collapse_tree`功能，可以将`collapse_tree`参数设置为`False`。禁用方法如下： > > response = RA.answer_question(question, collapse_tree=False) Then what is the point of building a tree? After you build the tree, do you search each node individually?
It is a bug and the auther dont fix it，so bad！！！！ and looking my test: https://github.com/VikParuchuri/marker/issues/593
> 服务器有3张gpu卡，通过如下命令docker部署指定其中一张显卡，显卡：  > > docker run --gpus '"device=2"' -d -e PROCESSES_PER_GPU=1 -p 7231:7231 --name model_pdf_v1 crpi-h3snc261q1dosroc.cn-hangzhou.personal.cr.aliyuncs.com/marker11/marker_images:latest > > 测试：curl --location --request POST "http://10.4.7.111:7231/v1/parse/file" --form "file=@/24v5.pdf" 通过上述命令测试了报错： {"detail":"错误信息: A child...
试试 image: registry.cn-hangzhou.aliyuncs.com/fastgpt/mongo:5.0.18 这个镜像
感谢你的贡献，可否代码优化一下，提个PR，放到plugins/model下
https://github.com/labring/FastGPT/pull/5662 可以看这个fastgpt接入的docker 镜像包，支持多个文件处理
最新版他有点问题 https://github.com/VikParuchuri/marker/issues/593 所以目前还是1.5.5版本的