info-nce-pytorch icon indicating copy to clipboard operation
info-nce-pytorch copied to clipboard

PyTorch implementation of the InfoNCE loss for self-supervised learning.

Results 4 info-nce-pytorch issues
Sort by recently updated
recently updated
newest added

I'm confused about this. NCEloss can determine the lower bound of mutual information. In this implementation, should NCEloss be minimized in order to increase mutual information?

Hi, Thanks very much for your work. I wonder how should I modify your loss to adapting to the case of multiple positive samples per query. For example, query.shape =...

I have a quick question. My data is (data1, data2, label), where the label indicates if the pair of data points represent the same thing. 0 being not the same...

Hi, I’m working with a set of prototypes that have the shape [b, num_cls, num_tokens, dim]. My goal is to use InfoNCE loss to maximize inter-class differences. I have the...