chibohe/text_recognition_toolbox

predict failed for DAN model

Johnson-yue opened this issue · 4 comments

Johnson-yue commented

hi， I trained the DAN model and got good performance ， but when I predicted with pretrained model
there is a error :
forward() missing 1 required positional augument: 'text'

And I check it in networks/DAN.py in line 42 :

why perdict need target text ??

Johnson-yue commented

Hi, I think the code for testing part have some bugs
such as :

  1. not set num_step variable in Line 298
  2. not used one_hot variable in Line 301
  3. not set prev_emb variable in Line 302
  4. fail to get one_hot variable in Line 301
chibohe commented

hi， I trained the DAN model and got good performance ， but when I predicted with pretrained model
there is a error :
forward() missing 1 required positional augument: 'text'

And I check it in networks/DAN.py in line 42 :

why perdict need target text ??

The predict.py has only been tested on CRNN and FAN, you can read DAN source code and make some modifications.

Johnson-yue commented

where is source code for DAN ？
did you reference here ：https://github.com/Wang-Tianwei/Decoupled-attention-network

you don‘t have this parameter self.char_embeddings in training， maybe I have to training again。。。

chibohe commented

where is source code for DAN ？
did you reference here ：https://github.com/Wang-Tianwei/Decoupled-attention-network

you don‘t have this parameter self.char_embeddings in training， maybe I have to training again。。。

where is source code for DAN ？
did you reference here ：https://github.com/Wang-Tianwei/Decoupled-attention-network

you don‘t have this parameter self.char_embeddings in training， maybe I have to training again。。。

The predict script for DAN has been implemented, please try again.