huggingface / alignment-handbook

Robust recipes to align language models with human and AI preferences
https://huggingface.co/HuggingFaceH4
Apache License 2.0
4.75k stars 413 forks source link

How to finetune or lora on custom dataset #9

Open universewill opened 1 year ago

universewill commented 1 year ago

How to finetune or lora on custom dataset

lewtun commented 1 year ago

Hi @universewill we've now released the code and added some instructions for training on your own datasets here: https://github.com/huggingface/alignment-handbook/tree/main/scripts#fine-tuning-on-your-datasets

Let me know if you hit any issues!

lizhen-lizhen commented 3 weeks ago

你好@universewill我们现在已经发布了代码,并在此处添加了一些使用您自己的数据集进行训练的说明:https://github.com/huggingface/alignment-handbook/tree/main/scripts#fine-tuning-on-your-datasets

如果您遇到任何问题请告诉我!

具体的数据格式是啥?