630.50 a

Description: RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrieval is an article that implements a great model of how to use RAGs. The method itself abstracts clusters in a tree-like fashion to provide a summary. So level 0 is the raw level, close to the original, and each level provides an abstracted level of the clusters.

#600 #ML_Libraries_and_Implementation#630 #Machine_Learning_Frameworks#630.50 #LlamaIndex#630.50 a #RAPTOR_RAG

code: https://github.com/murphybread/RAPTOR_Chroma_Local

reference
https://llamahub.ai/l/llama-packs/llama-index-packs-raptor?from=
https://blog.gopenai.com/deep-dive-into-raptor-with-llamaindex-raptor-pack-8eb4fa32df30
https://github.com/run-llama/llama_index/tree/main/llama-index-packs/llama-index-packs-raptor

RAPTOR install

llamaindex-cli download-llamapack RaptorPack --download-dir ./raptor_pack

pip install llama-index-vector-stores-chroma
pip install llama-index-embeddings-huggingface
pip install llama-index
pip install llama-index-packs-raptor

Reference Display

===

sqlite file online viwer

https://sqliteviewer.app/#/chroma.sqlite3/table/collections/

chroma_db

cloudformation template
https://s3.us-west-1.amazonaws.com/public.mintplexlabs.com/chromadb-with-api-key.json
참고영상
https://www.youtube.com/watch?v=4ApW6ZwqW20&ab_channel=VitoASMR