How to finetune this model?

by catastropiyush - opened May 16

May 16

I wanted to start finetuning VLMs but I am not able to find a good resource for this. Can we finetune this on Google Colab's T4 GPU? How to manage different image sizes in the dataset? Thank you.

OleehyO

Owner May 16

•

edited May 16

Yes, you can use T4 to train TexTeller, please refer to TexTeller’s GitHub repository for more details.

For images of different sizes in the dataset, they will be resized to 448 * 448 while maintaining the aspect ratio (the extra parts will be padded).

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment