RAG Baselines about self-rag

[AkariAsai]

I've uploaded the script to run baseline LMs!
https://github.com/AkariAsai/self-rag/blob/main/retrieval_lm/run_baseline_lm.py
I'll add documentations to run baselines, but essentially, you just need to specify the model name, and pass the same input file as in the Self-RAG pipeline. For retrieval baseline, please use --mode retrieval --prompt_name "prompt_no_input_retrieval" option to trigger retrieval.

e.g., Llama2-7b (pre-trained)

python run_baseline_refactor.py \
--model_name meta-llama/Llama-2-7b-hf \
--input_file INPUT_FILE_SAME_AS_SELF_RAG \
 --max_new_tokens 100 --metric match \
--result_fp RESULT_FILE_PATH --task qa --mode retrieval --prompt_name "prompt_no_input_retrieval"

e.g., ChatGT (March)

python run_baseline_refactor.py \
--model_name gpt-3.5-turbo-0301 \
--input_file INPUT_FILE_SAME_AS_SELF_RAG \
--max_new_tokens 100 --metric match \
--result_fp RESULT_FILE_PATH \
 --task qa \
--api_key YOUR_OPEN_AI_API_KEY_FILE \
--mode retrieval --prompt_name "prompt_no_input_retrieval" 

For OpenAI API models, you also need to set organization key here: https://github.com/AkariAsai/self-rag/blob/main/retrieval_lm/run_baseline_lm.py#L12

[AkariAsai]

I close this issue now but let me know if you have any further questions!

[robertgshaw2-neuralmagic]

Thank you :)