关于如何微调PP-OCRV4的检测模型

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)

Apache License 2.0

38.99k stars 7.32k forks source link

关于如何微调PP-OCRV4的检测模型 #12009

Closed TheHonestBob closed 2 weeks ago

TheHonestBob commented 2 weeks ago

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：
版本号/Version：Paddle： PaddleOCR：问题相关组件/Related components：
运行指令/Command Code：
完整报错/Complete Error Message：微调时，使用的预训练模型和配置文件是什么我测试了ch_PP-OCRv4_det_train和ch_PP-OCRv4下的cml和student配置文件，都未能成功训练

zhangyubo0722 commented 2 weeks ago

具体训练命令和报错信息是什么呢

zhangyubo0722 commented 2 weeks ago

python tools/train.py -c configs/det/ch_PP-OCRv4/ch_PP-OCRv4_det_student.yml -o Global.pretrained_model=xxx

TheHonestBob commented 2 weeks ago

python tools/train.py -c configs/det/ch_PP-OCRv4/ch_PP-OCRv4_det_student.yml -o Global.pretrained_model=xxx

我刚才又重新看了一下，是我的问题，我之前按照这个命令训练完成后，预测时指定det_dir指定到rec_dir上去了，我关闭这个issues，非常感谢