Instructions to use facebook/vit-mae-huge with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use facebook/vit-mae-huge with Transformers:
# Load model directly from transformers import AutoImageProcessor, AutoModelForPreTraining processor = AutoImageProcessor.from_pretrained("facebook/vit-mae-huge") model = AutoModelForPreTraining.from_pretrained("facebook/vit-mae-huge") - Notebooks
- Google Colab
- Kaggle
Finetuning on downstream dataset with a arbitrary mask ratio
#2
by AlirezaSalehi99 - opened
I'm trying to fine-tune the pretrained MAE on a downstream dataset using the same masked image modeling objective, but I’d like to change the mask ratio for this fine-tuning phase. Is it possible?
Thanks in advance
Any comment is appreciated