axel588

Results 3 comments of axel588

@HReynaud What training time do you have for training on 10 000 images on text to image ?

@lucidrains I applied the attention layer at first, but even with a dimension of 8 ( very low yes ) and a batch of 1 it overflows my 24gb memory...

thanks @apapiu for your answer the main issue I have is with 16_16_latent_embeddings.npy I have no idea how to reproduce this kind of file, Not sure how to transform images...