dmuqlzhang
dmuqlzhang
thanks for your great works! and I have a question, during testing, Can I detect multiple classes simultaneously? or the query must one classes? for examples, if my target images...
thanks for your job. will you release a inference images demo file about FSOD?
 I got Nan when testing in fsod testset, can you help me
excuse me, I want to understand why not add label weight at Multilabelloss function? I think maybe use weighted cross entropy loss will achieve a better result?
train error like this, i trained on python2.7, can you help me fix it ? 