Training on own dataset #17

Open
opened 2025-10-14 17:14:31 -06:00 by navan · 0 comments
Owner

Originally created by @hua-ops on 9/18/2023

Hello!
I have a single character trajectory dataset, I processed it to IAM-onDB format and trained the model, then the result is not very good. Can you please tell me if the model training prefers to use a long sequence of sentences rather than single characters, or any parameter tuning suggestions?
Thank you!

*Originally created by @hua-ops on 9/18/2023* Hello! I have a single character trajectory dataset, I processed it to IAM-onDB format and trained the model, then the result is not very good. Can you please tell me if the model training prefers to use a long sequence of sentences rather than single characters, or any parameter tuning suggestions? Thank you!
Sign in to join this conversation.
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference: github/handwriting-synthesis#17
No description provided.