search

zh320 / realtime-semantic-segmentation-pytorch

PyTorch implementation of over 30 realtime semantic segmentations models, e.g. BiSeNetv1, BiSeNetv2, CGNet, ContextNet, DABNet, DDRNet, EDANet, ENet, ERFNet, ESPNet, ESPNetv2, FastSCNN, ICNet, LEDNet, LinkNet, PP-LiteSeg, SegNet, ShelfNet, STDC, SwiftNet, and support knowledge distillation, distributed training etc.
Apache License 2.0
78 stars 14 forks source link

Where is the code for using pre-trained weight #13

Closed zhy168 closed 1 month ago

zhy168 commented 1 month ago

Thanks a lot for your code! I used your code of BiseNetV2 to train for 500 epochs and found that it converged very quickly. Did you use pre-trained weights? I iddn'n find the code of load pre-trained weights

zh320 commented 1 month ago

Hi zhy,

In this repo, currently only those models whose encoders are ResNet or MobilenetV2 are using pre-trained weights. BiseNetV2 does not have encoder, so it was trained from scratch.

You may also refer to issue #11.

Best,