search

open-compass / BotChat

Evaluating LLMs' multi-round chatting capability via assessing conversations generated by two LLM instances.
Apache License 2.0
136 stars 6 forks source link

Could you provide the code to generate dialogues with customized (local) models? #10

Open wln20 opened 10 months ago

wln20 commented 10 months ago

Hi! I really love your work and want to use it in our own research. However, I could not find the codes that can generate dialogue data with customized models in the repository. I only found the code to evaluate the existed dialogue data you had already generated. So could you please kindly provide the code to generate dialogues with customized (local) models? Thanks a lot!

frankweijue commented 10 months ago

Hello, thank you for your interest in our work. The code for generating dialogue can be found in app.py. If you need to add a new local model, you can refer to the implementation in hf_chat.py. Feel free to ask if you have any further questions.

syizhao commented 2 weeks ago

Thanks for your work. Could you please provide the evaluation code?