Qingzhang1994
Qingzhang1994
When I extract 1-million distractors descriptors, I find that the revisitop1m.txt file contains more images than the number of images extracted from 100 tar.gz files ( I download the http://ptak.felk.cvut.cz/revisitop/revisitop1m/),...
Can you provide a source folder like[ this](https://github.com/qliu24/SAKE/tree/master/dataset/Sketchy/zeroshot1)? It's a little difficult for me to implement those
When I trained the code, I found that the function required by the following code were missing. Could you please provide the code to help me study? weights = torch.DoubleTensor(utils.class_weights(tr_clss))...