Richard Tseng

Results 9 issues of Richard Tseng

@baudm hi, thanks for your great work. I have trained the parseq model with my own dataset, the max length is 89, and the chardict is 95full.yaml. when i test...

请问如何从数据集中构建词表呢? 以及构建完成的词表中如何解决类别分布不均的问题呢? ![image](https://user-images.githubusercontent.com/20828244/231641893-e9a148f3-448e-43d4-a1cc-0eef415e6091.png) ![image](https://user-images.githubusercontent.com/20828244/231641928-5bf57772-096a-4e5d-81f0-a992ae3f9b9d.png)

when i try to train a thai model, there are some nan/inf accrued in DeformableTransformer. ![image](https://github.com/ViTAE-Transformer/DeepSolo/assets/20828244/d8581867-2d53-4a8e-876d-5bff60628923) and i check the out put of self.detection_transformer: the output is [nan, nan, nan,......]...

n_layers=opt['num_layers'], rnn_cell=opt['rnn_type'], rnn_dropout_p=opt['rnn_dropout_p']).cuda() KeyError: 'rnn_type'

Thanks for your work, could you provide the S2VT model ?

hi, 'Key Error' occurs when I go to the evaluation phrase. is there any other file needed? how should I fix this error? thank you! ![image](https://github.com/clovaai/units/assets/20828244/db047ef6-b0a7-44ab-a2aa-2828259e64d6)

KeyError: 'e2e_method', when i run the commond "python contextualizer.py "

Thanks for your great job! Is there any plan to release the code?