Thank you and a few doubts

[sayakpaul]

Hi @wangz10,

Thank you for putting together this repository and also the blog post. After going through the code, I had the following doubts. If you could help me clarify I would really appreciate it:

• While training the baseline classifier, did you train the network without normalizing the features? In typical image classification models, the normalization step is not often seen I would say.
• Did you normalize the features for both the encoder and the projector?
• If you could provide some high-level explanation of the max_margin_contrastive_loss function, it would be very helpful.

Thank you and looking forward to hearing from you.

[wangz10]

Hi @sayakpaul ,

• Input images are normalized to [0,1] by dividing 255.
• Yes, the outputs from encoder and the projector are both normalized to have unit norm.
• max_margin_contrastive_loss is a wrapper for tfa.losses.contrastive_loss. I just computed the pairwise Euclidean/cosine distance, retrieve the upper triangle, then feed it into tfa.losses.contrastive_loss.

[sayakpaul]

Thanks for providing the additional details @wangz10.

I am referring to this implementation of the baseline architecture where you used wherein it seems like you have also normalized the features within the network which is where my doubt.

Regarding max_margin_contrastive_loss, could you provide some details regarding your approach towards computing the contrastive pairs?