-
### Bug description
Using "doctr/references/recognition/train_tensorflow.py" on Google Colab creates an error when I use GPU acceleration. If I only use the CPU everything works just fine.
### C…
-
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：Windows 10
- 版本号/Version：Paddle：2.5.0 PaddleOCR：2.6.1.3 PaddleHub: 2.3.1 问题相关…
-
hi,
i am using ch_ppocr_mobile_v1.1_rec_train.tar model and exported to fluid model and wanted to convert to onnx model. but failes with below error:
paddle.__version__ = 1.8.5
Now, paddle2onnx…
-
**模型来自PaddleOCR**：https://github.com/PaddlePaddle/PaddleOCR#pp-ocr-20-series-model-listupdate-on-dec-15
**运行命令**：python tools/quant.py -c configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml
**…
-
![文字较多定位比较差](https://user-images.githubusercontent.com/14155359/86690580-a1002980-c03a…
-
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
在不同服务器使用不同cuda版本编译的paddlegpu运行文字检测时结果相差极大，模型使用的均为同一通用模型，字典也没变化
- 系统环境/System Environment：
- linux/ubun…
-
纯CPU机器
使用hub serving start -m chinese_ocr_db_crnn_server启动没有问题
使用hub serving start -m chinese_ocr_db_crnn_server --use_multiprocess --workers启动就一直报错：
File "D:\Program\anaconda3\lib\site-packages\re…
-
论文里面理论上不是SRN算法大大优于CRNN吗，怎么会产生如此奇怪的结果。700万数据均是平直的文本，无需进行弯曲文本纠偏。不仅SRN算法如此，在不需要进行弯曲文本纠偏的中英文数据集上，DAN算法（Decoupled Attention Network for Text Recognition）的最终字符准确率也低于CRNN算法，这到底是为什么啊，我都快疯了。
PS：在全英文的数据集（30万，文…
-
### Prerequisite
- [X] I have searched [Issues](https://github.com/open-mmlab/mmocr/issues) and [Discussions](https://github.com/open-mmlab/mmocr/discussions) but cannot get the expected help.
- [X] …
-
I use the following yaml file, I even spent a few days training it, but every time I change it, it doesn't give me any accuracy.Thank you for helping me, where is the problem with my work?
`Global:
…