关于自定义数据集训练几个问题
#14871
Replies: 2 comments
-
关于你提出的 PaddleOCR 自定义数据集训练问题,我来详细回答: 1. 微调时选择哪个蒸馏配置文件?
2. 自定义数据集(例如 CASIA-HWDB2.x)训练配置
3. 混合训练 vs 纯自定义数据训练
相关参考你可以参考以下官方讨论:
希望这些信息能帮助你顺利进行 PaddleOCR 训练与微调!🚀 Response generated by 🤖 feifei-bot | chatgpt-4o-latest |
Beta Was this translation helpful? Give feedback.
0 replies
-
如果只训练微调 英文数字模型,应该期于哪个数据集做训练? 目前有500多张新的英文字体, 训练参数怎么调整较好? |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
🔎 Search before asking
🐛 Bug (问题描述)
1 . 文本识别模块微调用[ch_PP-OCRv3_rec_distillation.yml]
还是用[ch_PP-OCRv4_rec_distillation.yml] 比较好
(比如手写中文数据集[CASIA-HWDB2.x])调优的的话,训练需要配置什么能保存原先推理模型功能,
只用自定义数据集调试。还是说需要原先训练ch_PP- OCRv3_rec_infer/ch_PP-OCRv4_rec_infer数据集+自
定义数据集比例混合训练才可以?
🏃♂️ Environment (运行环境)
release/2.10.0版本
🌰 Minimal Reproducible Example (最小可复现问题的Demo)
1 . 文本识别模块微调用[ch_PP-OCRv3_rec_distillation.yml]
还是用[ch_PP-OCRv4_rec_distillation.yml] 比较好
(比如手写中文数据集[CASIA-HWDB2.x])调优的的话,训练需要配置什么能保存原先推理模型功能,
只用自定义数据集调试。还是说需要原先训练ch_PP- OCRv3_rec_infer/ch_PP-OCRv4_rec_infer数据集+自
定义数据集比例混合训练才可以?
Beta Was this translation helpful? Give feedback.
All reactions