Open Astrosun opened 7 years ago
@Astrosun Have you solved this pro yet? I also want to know :)
@nyyznyyz1991 I found that faster-rcnn has its own snapshot algorithm, which can enable you to resume your training.
@Astrosun Can you give me some specific help?
@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 program will at least resume the program from the last stage, for example, you just terminate your program in the middle of stage two detection training 20000k iterations, it will start at least from the stage two PRN training when your resume, but will start from 0 iterations.
@Astrosun THX a lot. I have just found the situation you mentioned.
@nyyznyyz1991 I found that faster-rcnn has its own snapshot algorithm, which can enable you to resume your training.
brother post it here
Hi, Is there a way to resume training faster-rcnn in the middle of the process? I think it will be good if we can resume the process like what caffe does just in case the computer loses power in the middle of the training.
Anyone can help? Thanks!