-
我参考kie\predict_kie_token_ser.py文件生成了一个SerPredictor,其中的ser是我自己训练的,det和rec用的pretrained_en。
先生成了一个参数列表
parser = init_args()
args=parser.parse_args(['--kie_algorithm=LayoutXLM',
'--ser_model_dir=f…
-
Hi all,
Many thanks for your excellent work on the LayoutXLM paper.
I am reproducing LayoutXLM relation extraction fine-tuning on the FUNSD dataset. (Training in FUNSD and test on FUNSD)
The…
-
请提供下述完整信息以便快速定位问题/Please provide the following information to quickly locate the problem
- 系统环境/System Environment:Ubuntu 20
- 版本号/Version:Paddle:2.3 PaddleOCR:2.6 问题相关组件/Related components:pytho…
-
请提供下述完整信息以便快速定位问题/Please provide the following information to quickly locate the problem
- 系统环境/System Environment:
ubuntu20.04
- 版本号/Version:Paddle: PaddleOCR: 问题相关组件/Related components:
paddl…
-
Hi, I find that the label_pad_token id is different between layoutxlm and ernie layout in token classification. In layoutxlm, the label_pad_token id is -100 and will be ignored by pytorch loss functi…
-
vi-LayoutXlm预测与det任务的box不一致,vi-LayoutXlm下游任务会使得文本框粘连,请问这是什么原因?
例如:
![image](https://user-images.githubusercontent.com/61568958/209101298-ff1e9416-aa58-4754-9aab-60f3deb04667.png)
和单独使用检测任务
![image…
-
以前是win10正常运行,后来重装系统,重装了win11的不同版本的gpu环境就出现乱码问题
NVIDIA Quadro M620, device: 0, GPU Compute Capability: 5.0, Driver API Version: 11.7, Runtime API Version: 11.7,cuDNN Version: 8.5,cuda-toolkit-cuda_11.…
-
```
*** Aborted at 1519973684 (unix time) try "date -d @1519973684" if you are using GNU date ***
PC: @ 0x0 (unknown)
*** SIGFPE (@0x7f89bd8a840a) received by PID 46 (TID 0x7f8793992…
-
Hi all,
I tested PaddleOCR on [ICDAR2013-Challenge2_Training_Task3_Images_GT](https://rrc.cvc.uab.es/?com=downloads&action=download&ch=2&f=aHR0cHM6Ly9ycmMuY3ZjLnVhYi5lcy9kb3dubG9hZHMvQ2hhbGxlbmdlMl9U…
-
请提供下述完整信息以便快速定位问题/Please provide the following information to quickly locate the problem
- 系统环境 python3.7.14
- 版本号/Version:Paddle 2.4.1, PaddleOCR: release/2.6
- 运行指令
- ```python3 tools/export…