├── DPO.ipynb ├── DPO └── readme.md ├── FT Eval Function ├── Fine-Tuning ├── Fine_tuning.ipynb └── readme.md ├── Model.ipynb ├── README.md ├── RLAIF ├── README.md ├── data_generation.py ├── dpo_train.py ├── quantize.py ├── requirements.txt ├── rlaif_data.json └── train.py ├── data ├── deploy_model.py ├── distillation ├── requirements.txt ├── run.sh ├── src │ └── un.md └── train.py ├── dpo_pairs.jsonl ├── evaluation.py ├── make_pairs.ipynb ├── models └── .gitkeep ├── requirements.txt ├── reward_model.py └── train_healthcare_llm.py /DPO.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/easonwangzk/MedLLM/HEAD/DPO.ipynb -------------------------------------------------------------------------------- /DPO/readme.md: -------------------------------------------------------------------------------- 1 | 2 | -------------------------------------------------------------------------------- /FT Eval Function: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/easonwangzk/MedLLM/HEAD/FT Eval Function -------------------------------------------------------------------------------- /Fine-Tuning/Fine_tuning.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/easonwangzk/MedLLM/HEAD/Fine-Tuning/Fine_tuning.ipynb -------------------------------------------------------------------------------- /Fine-Tuning/readme.md: -------------------------------------------------------------------------------- 1 | 2 | -------------------------------------------------------------------------------- /Model.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/easonwangzk/MedLLM/HEAD/Model.ipynb -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/easonwangzk/MedLLM/HEAD/README.md -------------------------------------------------------------------------------- /RLAIF/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/easonwangzk/MedLLM/HEAD/RLAIF/README.md -------------------------------------------------------------------------------- /RLAIF/data_generation.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/easonwangzk/MedLLM/HEAD/RLAIF/data_generation.py -------------------------------------------------------------------------------- /RLAIF/dpo_train.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/easonwangzk/MedLLM/HEAD/RLAIF/dpo_train.py -------------------------------------------------------------------------------- /RLAIF/quantize.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/easonwangzk/MedLLM/HEAD/RLAIF/quantize.py -------------------------------------------------------------------------------- /RLAIF/requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/easonwangzk/MedLLM/HEAD/RLAIF/requirements.txt -------------------------------------------------------------------------------- /RLAIF/rlaif_data.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/easonwangzk/MedLLM/HEAD/RLAIF/rlaif_data.json -------------------------------------------------------------------------------- /RLAIF/train.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/easonwangzk/MedLLM/HEAD/RLAIF/train.py -------------------------------------------------------------------------------- /data: -------------------------------------------------------------------------------- 1 | https://drive.google.com/file/d/1j2Z_rf_YM1sAC7WAWAsGIDnJtixhpGFZ/view?usp=share_link 2 | -------------------------------------------------------------------------------- /deploy_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/easonwangzk/MedLLM/HEAD/deploy_model.py -------------------------------------------------------------------------------- /distillation/requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/easonwangzk/MedLLM/HEAD/distillation/requirements.txt -------------------------------------------------------------------------------- /distillation/run.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/easonwangzk/MedLLM/HEAD/distillation/run.sh -------------------------------------------------------------------------------- /distillation/src/un.md: -------------------------------------------------------------------------------- 1 | 2 | -------------------------------------------------------------------------------- /distillation/train.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/easonwangzk/MedLLM/HEAD/distillation/train.py -------------------------------------------------------------------------------- /dpo_pairs.jsonl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/easonwangzk/MedLLM/HEAD/dpo_pairs.jsonl -------------------------------------------------------------------------------- /evaluation.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/easonwangzk/MedLLM/HEAD/evaluation.py -------------------------------------------------------------------------------- /make_pairs.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/easonwangzk/MedLLM/HEAD/make_pairs.ipynb -------------------------------------------------------------------------------- /models/.gitkeep: -------------------------------------------------------------------------------- 1 | 2 | -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/easonwangzk/MedLLM/HEAD/requirements.txt -------------------------------------------------------------------------------- /reward_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/easonwangzk/MedLLM/HEAD/reward_model.py -------------------------------------------------------------------------------- /train_healthcare_llm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/easonwangzk/MedLLM/HEAD/train_healthcare_llm.py --------------------------------------------------------------------------------