Does it work with greyscale training images? about pytorch-unet HOT 2 CLOSED

Make sure n_channels=1 and that your images are loaded are greyscale and it should be ok.

from pytorch-unet.

I tried this but got an error haven't been able to debug.

Make sure n_channels=1 and that your images are loaded are greyscale and it should be ok.

My batch size 15
channels: 1 (grayscale images)
Batch SHAPE: (15, 224, 224)
train.py : net = UNet(n_channels=1, n_classes=1)

TLDR:

RuntimeError: Expected 4-dimensional input for 4-dimensional weight 64 1 3, but got 3-dimensional input of size [15, 224, 224] instead

Full traceback log:

File "ushdrcnn_train.py", line 346, in train_net
masks_pred = net(imgs)
File "/home/user/anaconda3/envs/torch/lib/python3.7/site-packages/torch/nn/modules/module.py", line 493, in call
result = self.forward(*input, **kwargs)
File "/media/unet/main.py", line 34, in forward
x1 = self.inc(x)
File "/home/user/anaconda3/envs/torch/lib/python3.7/site-packages/torch/nn/modules/module.py", line 493, in call
result = self.forward(*input, **kwargs)
File "/media/unet/unet_parts.py", line 32, in forward
x = self.conv(x)
File "/home/user/anaconda3/envs/torch/lib/python3.7/site-packages/torch/nn/modules/module.py", line 493, in call
result = self.forward(*input, **kwargs)
File "/media/unet/unet_parts.py", line 22, in forward
x = self.conv(x)
File "/home/user/anaconda3/envs/torch/lib/python3.7/site-packages/torch/nn/modules/module.py", line 493, in call
result = self.forward(*input, **kwargs)
File "/home/user/anaconda3/envs/torch/lib/python3.7/site-packages/torch/nn/modules/container.py", line 92, in forward
input = module(input)
File "/home/user/anaconda3/envs/torch/lib/python3.7/site-packages/torch/nn/modules/module.py", line 493, in call
result = self.forward(*input, **kwargs)
File "/home/user/anaconda3/envs/torch/lib/python3.7/site-packages/torch/nn/modules/conv.py", line 338, in forward
self.padding, self.dilation, self.groups)
RuntimeError: Expected 4-dimensional input for 4-dimensional weight 64 1 3, but got 3-dimensional input of size [15, 224, 224] instead
`

from pytorch-unet.