assets/large_language_models/rag/components/create_faiss_index/spec.yaml

$schema: https://azuremlschemas.azureedge.net/latest/commandComponent.schema.json
type: command

tags:
    Preview: ""

version: 0.0.78
name: llm_rag_create_faiss_index
display_name: LLM - Create FAISS Index
is_deterministic: true

description: |
  Creates a FAISS index from embeddings. The index will be saved to the output folder.
  The index will be registered as a Data Asset named `asset_name` if `register_output` is set to `True`.

inputs:
  embeddings:
    type: uri_folder
    mode: direct
    description: Folder containing embeddings to be indexed.

outputs:
  index:
    type: uri_folder
    description: "Uri Folder containing MLIndex yaml file and faiss/pickle files that hold data for the index"

environment: azureml:llm-rag-embeddings@latest
code: '../src/'

command: >-
  python -m azureml.rag.tasks.build_faiss
  --embeddings '${{inputs.embeddings}}'
  --output '${{outputs.index}}'