is the checkpoint right?

[learn01one]

hi,i met a problem when ran the demo:
size mismatch for rec_head.decoder.tgt_embedding.weight: copying a param with shape torch.Size([91, 256]) from checkpoint, the shape in current model is torch.Size([72, 256]).
any detaills i don't get?thanks

[Shangbang Long]

Which script specifically did you run?

[learn01one]

Which script specifically did you run?

hi,the script:
./Experiment/Experiment2/Exper_2_STN_real_15_test.sh

the pretrained_model is :Real_15.pth.tar

[Shangbang Long]

Just try to uncomment line 33-34:

ICDAR2019-ArT-Recognition-Alchemy/Source/utils/labelmaps.py

Line 33 in 29e432e

 # special = '∑，ºóʃüεäèиíöλá：®é·＃' # high frequency in the public training set 

It may be that the model is the one we used in IC19 competition, which contains some rare characters.

[Shangbang Long]

Hi @learn01one
Did it help solve the problem?

[learn01one]

Hi @learn01one
Did it help solve the problem?
maybe,i met other problem:
RuntimeError: Error(s) in loading state_dict for ModelBuilder

[Shangbang Long]

Can you provide some more details?