search

donnyyou / torchcv

TorchCV: A PyTorch-Based Framework for Deep Learning in Computer Vision
https://pytorchcv.com
Apache License 2.0
2.25k stars 378 forks source link

[ERROR] DeepLabV3+DenseNet Backbone #44

Open stigma0617 opened 5 years ago

stigma0617 commented 5 years ago

Hi,

I have a question when I try to train DeepLabV3 with DenseNet backbone.

the loss term of DeepLabV3.

I wonder why you use this auxiliary loss term. As far as I know, original DeepLabV3 was not trained by the auxiliary cross-entropy loss.

As your code about deeplabv3 was hard-coded, it should be fixed for other backbone (i.g., DenseNet).

Thanks

donnyyou commented 5 years ago
  1. Original DeepLabV3 maybe not be trained with auxiliary, and I haven't tested the influence of it.
  2. The code is really hard-coded. You could simply fix it by replacing 2048 with self.backbone.num_features.
  3. Welcome your pull request!!!