dimension mismatch

[johnathanchiu]

I got this error

RuntimeError: Given groups=1, weight of size [512, 320, 1, 1], expected input[1, 2048, 1, 1] to have 320 channels, but got 2048 channels instead

I saw the other thread and I tried changing it to 512 but it still didn't work

[hangzhaomit]

Can you let me know the command you are running?

[johnathanchiu]

I'm running through PyCharm but here are my defaults:

[hangzhaomit]

please use the default fc_dim, which is 2048

[johnathanchiu]

I used the default. This is the new error I got:

RuntimeError: Error(s) in loading state_dict for PPMDeepsup:
size mismatch for cbr_deepsup.0.weight: copying a param with shape torch.Size([80, 160, 3, 3]) from checkpoint, the shape in current model is torch.Size([512, 1024, 3, 3]).
size mismatch for cbr_deepsup.1.weight: copying a param with shape torch.Size([80]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for cbr_deepsup.1.bias: copying a param with shape torch.Size([80]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for cbr_deepsup.1.running_mean: copying a param with shape torch.Size([80]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for cbr_deepsup.1.running_var: copying a param with shape torch.Size([80]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for cbr_deepsup.1._tmp_running_mean: copying a param with shape torch.Size([80]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for cbr_deepsup.1._tmp_running_var: copying a param with shape torch.Size([80]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for conv_last_deepsup.weight: copying a param with shape torch.Size([150, 80, 1, 1]) from checkpoint, the shape in current model is torch.Size([150, 512, 1, 1]).

[hangzhaomit]

What command are you running?

[johnathanchiu]

I'm not running any specific commands, I'm just using the defaults on pycharm. Should I run it from my terminal using the commands in the readme?

[hangzhaomit]

yes

[johnathanchiu]

I did everything from the readme. Still getting the same error not quite sure what's going wrong. I didn't change any dimensions. I'm running this bash command:

python3 -u test.py
--model_path /Users/[User]/Downloads/semantic-segmentation-pytorch-master-3/baseline-resnet50dilated-ppm_deepsup
--test_imgs /Users/[User]/Downloads/IMG_0106.jpeg
--arch_encoder resnet50dilated
--arch_decoder ppm_deepsup

[hangzhaomit]

If you are able to run the quick start ./demo_test.sh, then your setup should be correct.