想问一下关于ocr训练时的图像堆叠方式
#16579
Replies: 1 comment
-
|
您好，检测和识别可以直接去看PP-OCRv5 的配置文件，然后去PaddleOCR仓库搜素对应的数据增强函数。比如：PP-OCRv5_server_det 模型， 是经过各种数据增强操作后，随机crop出来640*640的区域参与训练。具体位置如下： |
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.
-
我没在源代码中找到，你们训练的时候比如，32个batch的文字图像，是直接resize再堆叠，还是等比例resize，padding完再堆叠的？如果可以能否贴一下源代码位置，或者我想修改在哪里修改比较好，感谢！
Beta Was this translation helpful? Give feedback.
All reactions