-
-
在输入同样的图片的时,几次测试结果不一致,时好时坏,请问是什么原因导致的呢?有什么方法可以使结果稳定吗?
-
I have scanned documents, pictures of document taken by phone etc. where first I need to detect single page to eliminate rest of image or fix rotation. Is it possible to extend a model by such a featu…
-
## 问题描述
请对问题进行描述,并提供日志或截图
**本项目不处理网络环境引发的问题**(例如 Empty translation result/Connection reset)
D:\>set DEEPL_SERVER_URL=https://deeplx.papercar.top/translate
(para) D:\>pdf2zh geo.pdf -s deepl
C:…
-
ckpt = torch.load(file, map_location="cpu")
1%|█ | 8/583 [00:04
-
return ssl_context.wrap_socket(sock)
File "D:\ProgramData\Anaconda3\lib\ssl.py", line 500, in wrap_socket
return self.sslsocket_class._create(
File "D:\ProgramData\Anaconda3\lib\ssl.py"…
-
我们尝试了大概50份文件,发现了一些存在的现象:
1、存在信息内容粘连情况。换行、空格等识别效果不太好
2、复杂结构容易出现内容错乱情况,比如左右结构的文档,比如:
文件内容:
![image](https://github.com/user-attachments/assets/57113364-a605-43ab-989f-d27bf451193b)
识别后模块:
![image]…
-
Traceback (most recent call last):
File "", line 198, in _run_module_as_main
File "", line 88, in _run_code
File "C:\anaconda\envs\ppp\Scripts\pdf2zh.exe\__main__.py", line 7, in
File "C:…
-
### Description of the bug | 错误描述
我的系统是centos7,所以下载的是magic-pdf[full,old_linux]版本,下载过程和cpu运行demo正常,在使用cuda加速时报了以下错误,将libcudnn_cnn_infer.so.8和libcudnn_ops_infer.so.8添加到LD_LIBRARY_PATH中也无济于事。
`reloca…
-
### Description of the bug | 错误描述
[11/14 10:44:21 d2.checkpoint.detection_checkpoint]: [DetectionCheckpointer] Loading from /home/gaoyuan/.cache/modelscope/hub/opendatalab/PDF-Extract-Kit-1___0/model…