Hey! Would you be able to provide some code for the Uniformity and Alignment metrics? Because I don't see you normalizing any embeddings during training -- do you do that after training only as you compute the metric?
We followed the original paper's code and did normalize for that evaluation. In our training we use cosine similarity, which also normalizes the embeddings.
Hey! Would you be able to provide some code for the Uniformity and Alignment metrics? Because I don't see you normalizing any embeddings during training -- do you do that after training only as you compute the metric?
Thanks!