search

PengyiZhang / SlimYOLOv3

This page is for the SlimYOLOv3: Narrower, Faster and Better for UAV Real-Time Applications
1.14k stars 320 forks source link

About train step #5

Open zhyj3038 opened 5 years ago

zhyj3038 commented 5 years ago

I wonder why step 1 and step 4 involve darknet, the pytorch version yolov3 can not wrk out it ?

PengyiZhang commented 5 years ago

@zhyj3038 Of course it works. We use Darknet just for the convenience of comparison with the original YOLOv3. You can also implement all the procedure of SlimYOLOv3 in pytorch for your convenience.

muyouyang commented 4 years ago

normal training is not neccessary for sparsity training? if i just want to prune the network can i skip the normal training step?