search

FishYuLi / BalancedGroupSoftmax

CVPR 2020 oral paper: Overcoming Classifier Imbalance for Long-tail Object Detection with Balanced Group Softmax.
Apache License 2.0
351 stars 64 forks source link

Question about the sample out in RFS #9

Open cnnAndBn opened 4 years ago

cnnAndBn commented 4 years ago

hi, I have a question about the implementation detail in RFS , in the class DistributedGroupSampler_addrepeat_sampleout defined in sampler.py file, there is a hard coding self.num_to_sample_out= [6000, 17000] , I read the code , it seems that you randomly exclude some images which need not to repeat resampling more than once and the exclusion number are 6000 and 17000 respectively for different group flag
image

if it is ,can you explain the reason why it is needed to do this exlusion and how to decide the exclusion number? thanks a lot

FishYuLi commented 4 years ago

@dadada101 Sorry for the confusion. Actually, DistributedGroupSampler_addrepeat_sampleout is not our implementation for RFS. It's just one of our tryings that does not work. (We should have deleted this class in sample.py) For our implementation of RFS, please refer to DistributedGroupSampler_addrepeat .

cnnAndBn commented 4 years ago

ok ，thanks a lot