eeric
eeric
The ResNet model pruned impletement faster little about 10%, code in https://github.com/eeric/channel_prune the reason that non-tensor layers (e.g., batch normalization and pooling layers) took up more than 40% of the...
test myself
@buaaJeremyduan, thanks!
ResNet18 model pruned successfully,code in https://github.com/eeric/channel_prune
@zhenheny, thanks, so I didn't pay attention to it at present.
did not use, you could try as following: https://github.com/eeric/yolo2voc2coco
@milulee,@ShirelJosef you debuged fine_tuner.prune() on the 360th line in finetune.py, and then debug def prune(self): on the 278th line in finetune.py.
to my model prunned, it was lower than original by 1.5% for accuracy.
ok, maybe it was not excellent to your original model.