Astrosun

@nyyznyyz1991 I found that faster-rcnn has its own snapshot algorithm, which can enable you to resume your training.

@Drumaddheart, I did not look at too much details of the snapshot. But what I found is that if you terminate your program in the middle of the process, the...

hello, have you solved this problem? I have the same problems when running experiments/script_faster_rcnn_VOC2007_ZF.m. Thanks~

@duygusar May I know how you deal with the image mean part? So for my own data set, do we need to compute the image means of our own images?...

@duygusar Thanks so much for your valuable advice and it will be the greatest help! By the way, one more question, when you calculate the mean of your training data...

@duygusar Thanks a lot for your valuable advice and I will try it!

@zhihaolee @hqqxyy @ShaoqingRen I am having the same problem with you. I run out of cpu memory during training my own dataset in the stage one's do_proprosal_test() function which tries...

@xiaoxiongli , Hi, may I know how you save memory (reduce to 10%) in function do_proprosal_test or how you put the nms and subset - selection operations into the proposal_test...

@xiaoxiongli ,Thanks a lot for your help!