Just asking. Is non-english chars allowed?

Question

Just asking. Is non-english chars allowed?

muhammedcanpirincci-sudo opened this issue 3 years ago · comments

Muhammed Can Pirinççi commented 3 years ago

Can i add my language's char's to charList.txt and train with my language's dataset?
Thanks.

Harald Scheidl · Answer 1 · Sat Dec 04 2021 00:04:36 GMT+0800 (China Standard Time)

you have to rewrite the data loader for your dataset and then train the model. the chars file is automatically created.
some hints on how to do this see: https://towardsdatascience.com/faq-build-a-handwritten-text-recognition-system-using-tensorflow-27648fb18519

Harald Scheidl · Answer 2 · Sat Dec 04 2021 00:22:09 GMT+0800 (China Standard Time)

so you basically have to rewrite this class so that it loads your dataset but still has the same interface as it has now:
https://github.com/githubharald/SimpleHTR/blob/master/src/dataloader_iam.py

Muhammed Can Pirinççi · Answer 3 · Sat Dec 04 2021 00:23:18 GMT+0800 (China Standard Time)

so this part. : 1.2 Create IAM-compatible dataset and train model". I saw it but i just wanted to be sure. Thank you so much for your time sir.

Harald Scheidl · Answer 4 · Sat Dec 04 2021 00:25:31 GMT+0800 (China Standard Time)

yes, either make your dataset look like the IAM dataset and use the original dataloader.
Or write a new dataloader that loads your dataset and make the dataloader look like the original one.

Whatever you prefer.