Is there any way to change sparse tensor to full tensor?

I am new to pytorch. I have a sparse tensor and a full one and I want to find the similarity between these two binary tensors. I tried binary_cross_entropy_with_logits but it isn’t for a sparse and a full tensor and I couldn’t find anything else, so I tried to change sparse tensor to full, and tried to map sparse tensor to full one by casting (torch.FloatTensor), but it doesn’t work out. What can I do?


1 Like