graphrag/tests/integration/_pipeline/megapipeline.yml

input:
  file_type: text
  base_dir: ../../fixtures/min-csv
  file_pattern: .*\.txt$

storage:
  type: memory

cache:
  type: memory

workflows:
  - name: create_base_text_units
    config:
      text_chunk:
        strategy:
          type: sentence

      # Just lump everything together
      chunk_by: []

  - name: create_base_extracted_entities
    config:
      graphml_snapshot: True
      entity_extract:
        strategy:
          type: nltk

  - name: create_base_entity_graph
    config:
      graphml_snapshot: True
      embed_graph_enabled: True
      cluster_graph:
        strategy:
          type: leiden
          verbose: True

  - name: create_final_nodes

  - name: create_base_documents

  # - name: create_final_community_reports
  #   config:
  #     create_community_reports:
  #       <<: *llm_parallel_config
  #       strategy:
  #         type: graph_intelligence
  #         llm: *llm_config

  - name: create_final_communities
  - name: create_final_text_units
    config:
      text_embed:
        strategy:
          type: mock

  - name: create_final_entities
    config:
      text_embed:
        strategy:
          type: mock

  - name: create_final_documents
    config:
      text_embed:
        strategy:
          type: mock
  - name: create_final_relationships
    config:
      text_embed:
        strategy:
          type: mock