usuyama / pytorch-unet

Simple PyTorch implementations of U-Net/FullyConvNet (FCN) for image segmentation
https://colab.research.google.com/github/usuyama/pytorch-unet/blob/master/pytorch_unet_resnet18_colab.ipynb
MIT License
840 stars 233 forks source link

RGB Target mask #17

Open AndranikUgujyan opened 2 years ago

AndranikUgujyan commented 2 years ago

Firstly thanks for your implantation, it is really so good.

Now I want to try if the input image is grayscale [512, 512] and the generated target mask is RGB image [512, 512, 3] (same figures). After SimDataset and DataLoader we have shapes torch.Size([20, 512, 512]) torch.Size([20, 512, 512, 3]) and training show RuntimeError: Given groups=1, weight of size [64, 3, 3, 3], expected input[1, 20, 512, 512] to have 3 channels, but got 20 channels instead.

How to fix this issue?