MLE and hMup code

[ddy-ddy]

Hello, author, I like your article very much, and I plan to carry out scientific research on the basis of the technology you mentioned in this article. Now I'm replicating your experiment. I wonder if you can provide the code and parameters of MLE and HMUP that you used in the experiment.
Thank you very much.
Here is my email:1179730251@qq.com

[blmoistawinde]

Sorry I couldn't find my exact experiment code for them, since it's been a while😂 But these exps are mainly based on the following reporsitory:
MLE is from https://github.com/raosudha89/clarification_question_generation_pytorch , but you may need to change some codes in main.py, decode.py, evaluate.py with mine here as the original code contains bug.
hMup is from https://github.com/pytorch/fairseq/tree/master/examples/translation_moe , but you may need to change the architecture to LSTM and adjust some hyperparameters for a fair comparison.

[ddy-ddy]

Sorry I couldn't find my exact experiment code for them, since it's been a while😂 But these exps are mainly based on the following reporsitory:
MLE is from https://github.com/raosudha89/clarification_question_generation_pytorch , but you may need to change some codes in main.py, decode.py, evaluate.py with mine here as the original code contains bug.
hMup is from https://github.com/pytorch/fairseq/tree/master/examples/translation_moe , but you may need to change the architecture to LSTM and adjust some hyperparameters for a fair comparison.

ok. thank you so much.