Grayscale images in u-net

vision
3

Hi Ryan,

Thanks for your reply! I was trying to look to your code and the example you linked. I’ve been looking to your post here (I quote the original)

So I tried to modify also the first layer in the Unet I’m using by writing:
self.dconv_down1 = double_conv(1, 64)

Before there was 3 being the three channels RGB I guess.

I’ve tried to run the training and I get this error now

invalid argument 0: Sizes of tensors must match except in dimension 1. Got 263 and 272 in dimension 2 at C:/w/1/s/tmp_conda_3.7_055457/conda/conda-bld/pytorch_1565416617654/work/aten/src\THC/generic/THCTensorMath.cu:71

I’ve checked all the images they are 1024x1024 png grayscale images but I’m not sure how to solve this issue.

Best