MedGPT

Code related to work on - Reinforcement Learning from Human Feedback as an Alternative to Create Domain Expert A Case Study on Medical LLMs

Name		Name	Last commit message	Last commit date
Latest commit History 9 Commits
datasets		datasets
model_weights		model_weights
ui		ui
README.md		README.md
Report.pdf		Report.pdf
eval_bart_score.py		eval_bart_score.py
eval_ppo_output.py		eval_ppo_output.py
gemini_evaluator.py		gemini_evaluator.py
generate_reward_data.py		generate_reward_data.py
ppo_training.py		ppo_training.py
reward_modeling_gpt2.py		reward_modeling_gpt2.py
sft_training.py		sft_training.py
utility.py		utility.py

Provide feedback