Cohere Ranking
vector-searchsimple-rerankingvector-databaseretrieval-augmented-generationllm-frameworksweaviate-featurescohere-rankingfunction-callingweaviate-recipesPythongenerative-aireranking
Export
Dependencies
[ ]
Configuration
[1]
embedded weaviate is already listening on port 8079
Create a collection
Collection stores your data and vector embeddings.
[ ]
Upload Data
[6]
Manually chunk up the document into smaller chunks. This results in the chunks being a bit messy, but this can be improved by using an external tool like LlamaIndex, Haystack, LangChain, etc.
[7]
Query
Query without reranking
[ ]
The first few results from the above query aren't exactly what we're looking for. Let's run the query again, but rerank the top 10 documents with the text in the content property.
Query with Ranking
[ ]