PaddlePaddle / PaddleOCR

Awesome multilingual OCR toolkits based on PaddlePaddle (practical ultra lightweight OCR system, support 80+ languages recognition, provide data annotation and synthesis tools, support training and deployment among server, mobile, embedded and IoT devices)
https://paddlepaddle.github.io/PaddleOCR/
Apache License 2.0
42.44k stars 7.66k forks source link

./configs/rec/rec_svtrnet_ch.yml 这个配置可以训练 ./configs/rec/PP-OCRv4/ch_PP-OCRv4_rec.yml 这个配置即使设置batch size =1也爆显存,怎么修改这个配置让其能训练? #12584

Closed nissansz closed 3 months ago

nissansz commented 3 months ago

问题描述 / Problem Description

运行环境 / Runtime Environment

复现代码 / Reproduction Code

完整报错 / Complete Error Message

可能解决方案 / Possible solutions

附件 / Appendix

./configs/rec/rec_svtrnet_ch.yml 这个配置可以训练 ch_PP-OCRv4_rec_svtr_large4lan.yml 这个配置也可以训练

./configs/rec/PP-OCRv4/ch_PP-OCRv4_rec.yml 这个配置即使设置batch size =1也爆显存,怎么修改这个配置让其能训练?

GreatV commented 3 months ago

升级一下paddle版本、换一个更大显存的服务器

nissansz commented 3 months ago

哪个svtr的配置得到的模型准确率最高?