Al-Hitawi Mohammed

Results 15 comments of Al-Hitawi Mohammed

@caihaibin1991 1- First clone you need to clone the repo `git clone https://github.com/Mohammed20201991/TextRecognitionDataGeneratorHuMu23.git` or you can clone the original one made by cc @Belval 2- `pip install requirements.txt` 3- `cd...

@nag1045 You can see I am adding new language her [add new lang](https://github.com/Mohammed20201991/TextRecognitionDataGeneratorHuMu23)

cc @Belval can you please take a look at this issue ? thanks in advance

For the second part of generating **Hungarian** and **English** synthetic handwritten data best tool I found [TRDGHuMo](https://github.com/Mohammed20201991/TextRecognitionDataGeneratorHuMu23)

I do not understand what happens when running the `accelerate test` command when I am using `A100 (80GB) GPUs` ``` Copy and paste the text below into your GitHub issue...

Thanks to **Natabara** for his comment The solution is super easy by just skipping start token `labels= labels[1:] ` coming from the tokenizer because the tokenizer adds start token and...