Integrated Perception Models to my features extraction loss function.

I just updated the feature extraction code I use in some of my projects to work with the Perception Models. It probably still needs some work, but I'm sharing it here in case it's useful to anyone.

Here's a quick example I used to check if the features were being extracted correctly:


Link: [https://github.com/BurguerJohn/RepAlignLoss/blob/main/RepAlignLoss.py](https://github.com/BurguerJohn/RepAlignLoss/blob/main/RepAlignLoss.py)