Explain, analyze, and visualize NLP language models. Ecco creates interactive visualizations directly in Jupyter notebooks explaining the behavior of Transformer-based language models (like GPT2, BERT, RoBERTA, T5, and T0).
Hi, I am wondering how to serialize the LMOutput results for later usage?
I think it would be usefule when I want to explain the results of the whole dataset and visualize them later.
I have tried pickle.dump but meet this error "AttributeError: Can't pickle local object 'add_hook_to_module..new_forward'".
Hi, I am wondering how to serialize the LMOutput results for later usage? I think it would be usefule when I want to explain the results of the whole dataset and visualize them later. I have tried pickle.dump but meet this error "AttributeError: Can't pickle local object 'add_hook_to_module..new_forward'".
Thanks a lot!