search

YuzheZhang-1999 / DiffTSR

[CVPR2024] Diffusion-based Blind Text Image Super-Resolution (Official)
67 stars 4 forks source link

RuntimeError #9

Open 8600862 opened 2 months ago

8600862 commented 2 months ago

just run python inference_DiffTSR.py RuntimeError: Error(s) in loading state_dict for Transformer: size mismatch for embedding_word.lut.weight: copying a param with shape torch.Size([7937, 512]) from checkpoint, the shape in current model is torch.Size([23558, 512]). size mismatch for generator_word.proj.weight: copying a param with shape torch.Size([7937, 1024]) from checkpoint, the shape in current model is torch.Size([23558, 1024]). size mismatch for generator_word.proj.bias: copying a param with shape torch.Size([7937]) from checkpoint, the shape in current model is torch.Size([23558]).

YuzheZhang-1999 commented 1 month ago

Thank you for your attention and interest. This error is quite unusual because, as long as the model configuration file hasn't been modified, a loading error shouldn't occur. Could you have made changes to some parts of the code? If you could provide more information, it would help me assess the situation.