Replies: 1 comment
-
Vanilla SVTR does not support variable length training, but PPOCR-v3 yes. |
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
-
Hi Everyone,
I'm using Vanilla SVTR to train recognition.
And wanted to train it on sentences of varying length, but didn't know if that was possible using the current arrangement of the model which might be different than PPOCR-v3
What I know is that Self-Attention + CTC have a great performance in capturing long sentences
and I found that it has CTC as its head
I'm thrilled to know what you think about this?
Beta Was this translation helpful? Give feedback.
All reactions