dmuqlzhang

i think this is a bug，for example， 10 shots. refer the paper config(novel k shot, base 3×k shot), about 15 base classes, i should shoose 15 base classes( 15×10×3); about...

> OK, I will send the link in 3 days to you (I need to download the translated images from server and then upload them to netdisk, which may take...

thanks for your reply.  but I got None about this in voc_eval.py, I think predict image id not equal GT image id , is right？

> thanks for your reply. > >  > > but I got None about this in voc_eval.py, I think predict image id not equal GT image id , is...

> your trying to reshape "cls" to 64x15x845, but your "cls" is 1081600=64x20x845. Checkout your "num_anchors" and "num_classes" which should be 5 and 1 respectively. Or checkout the filter number...

> I'm not sure, but maybe your "meta loader" is using full 20 classes, instead of 15 base classes. Class MetaDataset prints out number of classes like below. If it...

thanks for your quick reply! i check your advice like this:   should I delete othor 15 base class only retain 5 novel in voc_novels.txt?

can I check your network structure?   1081600 = 1280 * 5 * 13 * 13 , what meaning of 1280?

> Hmm... strange... :( > Given dataloader automatically select base classes according to voc_novel.txt so you don't need to erase anything. > Output of layer 29 is concat of layer...

> I get 960x5x13x13 for cls shape before cls.view, which is same value as 64x15x845. > And you don't need to delete novel classes in voc_traindict_full.txt > > In dynamic...