Closed hktxt closed 5 years ago
@hktxt this optionally freezes the darknet53 backbone for the first epoch so only non-backbone layers are trained.
with open()
commands don't require close()
commands.
Hi @glenn-jocher ,
It would be really great to hear your comments! Thanks
@Pari-singh --transfer freezes layers, which I do not recommend. Simply train normally for best results from whatever starting point you wish using python3 train.py --weights start_from_here.pt
@glenn-jocher which eventually means training it more on entire model, correct? I was wondering if there was a feature for gradual unfreezing till the layer we want
@Pari-singh no there's no feature like this. If you want the fastest results, simply start from pretrained yolov3-spp.
python3 train.py --weights yolov3-spp.weights
If you want the best results, train on randomly initialized weights:
python3 train.py --weights ''
Yes, thanks for the insight @glenn-jocher
You're welcome, @Pari-singh! If you need further assistance, feel free to ask. Good luck with your training!
Hey, thanks for your implementation. I just don't understand this:
# Freeze backbone at epoch 0, unfreeze at epoch 1
in train loop.By the way, no close() were called, while using open(). I think switch it to with open() is more elegant. And, add a param for training from scratch, even though we can do this just comment some lines.