Cityscapes evaluation

[Vaportoxic1]

Hello Lavreniuk

Which backbone did you on cityscapes

n2.bias", "encoder.layer3.1.bn2.running_mean", "encoder.layer3.1.bn2.running_var", "encoder.layer4.0.conv1.weight", "encoder.layer4.0.bn1.weight", "encoder.layer4.0.bn1.bias", "encoder.layer4.0.bn1.running_mean", "encoder.layer4.0.bn1.running_var", "encoder.layer4.0.conv2.weight", "encoder.layer4.0.bn2.weight", "encoder.layer4.0.bn2.bias", "encoder.layer4.0.bn2.running_mean", "encoder.layer4.0.bn2.running_var", "encoder.layer4.0.downsample.0.weight", "encoder.layer4.0.downsample.1.weight", "encoder.layer4.0.downsample.1.bias", "encoder.layer4.0.downsample.1.running_mean", "encoder.layer4.0.downsample.1.running_var", "encoder.layer4.1.conv1.weight", "encoder.layer4.1.bn1.weight", "encoder.layer4.1.bn1.bias", "encoder.layer4.1.bn1.running_mean", "encoder.layer4.1.bn1.running_var", "encoder.layer4.1.conv2.weight", "encoder.layer4.1.bn2.weight", "encoder.layer4.1.bn2.bias", "encoder.layer4.1.bn2.running_mean", "encoder.layer4.1.bn2.running_var".
size mismatch for decoder.blocks.0.conv1.conv.weight: copying a param with shape torch.Size([1024, 4224, 3, 3]) from checkpoint, the shape in current model is torch.Size([1024, 768, 3, 3]).
size mismatch for decoder.blocks.1.conv1.conv.weight: copying a param with shape torch.Size([512, 1728, 3, 3]) from checkpoint, the shape in current model is torch.Size([512, 1152, 3, 3]).
size mismatch for decoder.blocks.2.conv1.conv.weight: copying a param with shape torch.Size([256, 864, 3, 3]) from checkpoint, the shape in current model is torch.Size([256, 576, 3, 3]).
size mismatch for decoder.blocks.3.conv1.conv.weight: copying a param with shape torch.Size([128, 256, 3, 3]) from checkpoint, the shape in current model is torch.Size([128, 320, 3, 3]).

I am getting mismatch trying to evaluate your pretrained model from huggingface. thank you very much in advance

[Lavreniuk]

hi, @Vaportoxic1 , I used the biggest one, so you need to use this config:
https://github.com/Lavreniuk/SPIdepth/blob/main/args_files/args_cvnXt_H_cityscapes_finetune_eval.txt

did you use it, or different one?

[Vaportoxic1]

convnextv2_huge.fcmae_ft_in22k_in1k_384 this is the backbone I am passing

[Lavreniuk]

the backbone is correct that you used.
Do you run this command
python ./tools/evaluate_depth_cityscapes_config.py args_files/args_cvnXt_H_cityscapes_finetune_eval.txt
I just double check and it works fine...

[yyy-123-abc]

你好 Lavreniuk

你在城市景观方面的主干是哪位

n2.bias“， ”encoder.layer3.1.bn2.running_mean“， ”encoder.layer3.1.bn2.running_var“， ”encoder.layer4.0.conv1.weight“， ”encoder.layer4.0.bn1.weight“， ”encoder.layer4.0.bn1.bias“， ”encoder.layer4.0.bn1.running_mean“， ”encoder.layer4.0.bn1.running_var“， ”encoder.layer4.0.conv2.weight“， ”encoder.layer4.0.bn2.weight“， ”encoder.layer4.0.bn2.bias“， ”encoder.layer4.0.bn2.running_mean“， ”encoder.layer4.0.bn2.running_var“， ”encoder.layer4.0.downsample.0.weight“， ”encoder.layer4.0.downsample.1.weight“， ”encoder.layer4.0.downsample.1.bias“， ”encoder.layer4.0.downsample.1.running_mean“， ”encoder.layer4.0.downsample.1.running_var“， ”encoder.layer4.1.conv1.weight“， ”encoder.layer4.1.bn1.weight“， ”encoder.layer4.1.bn1.bias“， ”encoder.layer4.1.bn1.running_mean“， ”encoder.layer4.1.bn1.running_var“， ”encoder.layer4.1.conv2.weight“， ”encoder.layer4.1.bn2.weight“， ”encoder.layer4.1.bn2.bias“， ”encoder.layer4.1.bn2.running_mean”， “encoder.layer4.1.bn2.running_var”。 decoder.blocks.0.conv1.conv.weight 的大小不匹配：使用 shape torch 复制参数。size（[1024， 4224， 3， 3]） 中，当前模型中的形状是 torch。大小（[1024， 768， 3， 3]）. decoder.blocks.1.conv1.conv.weight 的大小不匹配：使用 shape torch 复制参数。size（[512， 1728， 3， 3]） 中，当前模型的形状是 torch。大小 （[512， 1152， 3， 3]）. decoder.blocks.2.conv1.conv.weight 的大小不匹配：使用 shape torch 复制参数。size（[256， 864， 3， 3]） 中，当前模型的形状是 torch。大小（[256， 576， 3， 3]）. decoder.blocks.3.conv1.conv.weight 的大小不匹配：使用 shape torch 复制参数。size（[128， 256， 3， 3]） 的 checkpoint 中，当前模型的形状是 torch。大小 （[128， 320， 3， 3]）.

我在尝试评估来自 huggingface 的预训练模型时遇到不匹配。提前非常感谢

Hello, have you solved this problem? I'm experiencing the same issue.