search

lucidrains / vit-pytorch

Implementation of Vision Transformer, a simple way to achieve SOTA in vision classification with only a single transformer encoder, in Pytorch
MIT License
20.65k stars 3.05k forks source link

Removing FC layer at the top of the transformer model #164

Open loukasilias opened 3 years ago

loukasilias commented 3 years ago

Hi all, I would like to ask you how I can remove the fully connected layer at the top of the transformer layer. I want to get the output of the transformer networks and not the final prediction. Moreover, I would like to ask if these models come with the pretrained weights.

Thank you.

mghayour commented 2 years ago

I have same question on this, do we have any updates !?

mcihadarslanoglu commented 1 year ago

maybe this post can help you.