nadeeshamedagama/RAG Milvus Vision Pipeline

Production RAG app using Azure OpenAI, Milvus, LangGraph, SOLID principles.

View on GitHub

Trust Signals

Scorecard Score
not yet scored
Maintenance Recency
Stale
License
None
namedescriptionrequireddefault
milvus-uriMilvus Cloud cluster URIyes
milvus-tokenMilvus Cloud API tokenyes
milvus-collection-nameName of the Milvus collectionnorag_collection
azure-openai-api-keyAzure OpenAI API key for embeddingsyes
azure-openai-endpointAzure OpenAI endpoint URLyes
azure-openai-embedding-deploymentAzure OpenAI embedding deployment namenotext-embedding-ada-002
azure-openai-api-versionAzure OpenAI API versionno2024-02-01
embedding-dimensionDimension of the embedding vectorsno1536
github-repo-urlGitHub repository URL to process (optional)no""
data-directoryPath to local data directory with diagrams and documentsno./data/diagrams
google-vision-credentials-jsonGoogle Vision API service account JSON (base64 encoded)no""
google-vision-max-resultsMaximum results for Google Vision API analysisno10
skip-existingSkip documents that already exist in the vector storenotrue
force-reprocessForce reprocessing of all documentsnofalse
modeProcessing mode: ingest (run pipeline) or query (start API server)noingest
query-textQuery text for retrieval (only used in query mode)no""
top-kNumber of top results to return in query modeno5
namedescription
statusPipeline execution status (success/failed)
documents-processedNumber of documents processed
embeddings-createdNumber of embeddings created and stored
collection-nameName of the Milvus collection used
query-resultsQuery results in JSON format (only in query mode)