hillaryke/rag-optimization-cnn-dailymail
Optimizing a Retrieval-Augmented Generation (RAG) system on the CNN/Daily Mail dataset using LangChain, with performance benchmarking and analysis via RAGAS.
Jupyter NotebookMIT
Issues
- 0
- 0
- 0
- 0
- 0
- 0
- 0
- 0
Update README - Document my approach
#36 opened - 0
Implement the optimization technique.
#35 opened - 0
Identify optimization techniques to implement based effectiveness on the areas of weakness.
#34 opened - 0
- 0
- 0
Store the created embeddings to pgvector
#28 opened - 0
- 0
- 0
- 0
Setup pgvector - creating docker setup
#24 opened - 0
- 0
- 0
Choose the metrics for evaluation.
#20 opened - 0
Setup a RAGAS testing pipeline.
#19 opened - 0
- 0
- 1
- 0
install datasets python library
#13 opened - 0
- 0
- 3
- 0
Ensure poetry is installed
#9 opened - 0
2. Data Preparation and Setup
#8 opened - 0
3. Basic RAG system implementation
#7 opened - 0
Embed documents and store embeddings
#6 opened - 0
7. Re-benchmarking and analysis
#5 opened - 0
6. RAG Optimization
#4 opened - 0
- 0
4. Setup RAG retrieval mechanism
#2 opened - 0
1. Setup environment
#1 opened