alihan/KG_RAG
1
0
Fork 0
mirror of https://github.com/BaranziniLab/KG_RAG.git synced 2024-06-08 14:12:54 +03:00
Code Issues Packages Projects Releases Wiki Activity

adding quantitivate evidence to system prompt

Browse Source
This commit is contained in:
Karthik Soman
2024-03-13 14:36:58 -07:00
parent 760045c73a
commit 6639f77850
2 changed files with 2 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
kg_rag/rag_based_generation/GPT/drug_repurposing_v2.py
View File

@@ -35,7 +35,7 @@ print('')
question = input("Question : ")
question_template = f'''
To the question asked at the end, answer by referring the context. Reason with quantitative evidence.
To the question asked at the end, answer by referring the context.
See example below
Example 1:
Question:

1
system_prompts.yaml
View File

@@ -94,6 +94,7 @@ DRUG_REPURPOSING_V2: |
Then stitch these associations to give your final answer to the question.
More importantly, DO NOT consider 'Compound TREATS Disease' as the repurposing candidate.
Do not forget to give Provenance information which you can find from the context provided.
Reason with quantitative evidence, if possible (i.e. incuding statistical values if available).
For example:
Question:
What drugs can be repurposed to treat disease X?