.project-metadata.yaml

name: RAG Studio
description: "Build a RAG application to ask questions about your documents"
author: "Cloudera"
date: "2024-09-10"
specification_version: 1.0
prototype_version: 1.0

environment_variables:
    AWS_DEFAULT_REGION:
        default: "us-west-2"
        description: "AWS Region where Bedrock is configured and the S3 bucket is located. Required if Bedrock or S3 is used."
        required: false
    S3_RAG_DOCUMENT_BUCKET:
        default: ""
        description: "The S3 bucket where uploaded documents are stored. Only set if S3 is used for file storage."
        required: false
    S3_RAG_BUCKET_PREFIX:
        default: "rag-studio"
        description: "A prefix added to all S3 paths used by Rag Studio. Only recommended if S3 is used for file storage."
        required: false
    AWS_ACCESS_KEY_ID:
        default: ""
        description: "AWS Access Key ID. Required if Bedrock or S3 is used."
        required: false
    AWS_SECRET_ACCESS_KEY:
        default: ""
        description: "AWS Secret Access Key. Required if Bedrock or S3 is used."
        required: false
    USE_ENHANCED_PDF_PROCESSING:
        default: "false"
        description: "Use enhanced PDF processing for better text extraction. This option makes PDF parsing take significantly longer. A GPU is highly recommended to speed up the process."
        required: false
    CAII_DOMAIN:
        default: ""
        description: "The domain of the CAII service. Setting this will enable CAII as the sole source for both inference and embedding models."
        required: false
    DB_URL:
        default: "jdbc:h2:file:~/databases/rag"
        description: "Internal DB URL. Do not change."
        required: true
    LLM_SERVICE_URL:
        default: "http://localhost:8081"
        description: "Internal API URL. Do not change."
        required: true
    API_URL:
        default: "http://localhost:8080"
        description: "Internal API URL. Do not change."
        required: true


runtimes:
  - editor: JupyterLab
    kernel: Python 3.10
    edition: Standard

tasks:
  - type: run_session
    name: Download/Install Project Dependencies
    script: scripts/01_install_base.py
    short_summary: Install Project Dependencies
    kernel: python3
    cpu: 2
    memory: 4

  - type: create_job
    name: Update/build RAG Studio
    entity_label: refresh_project
    script: scripts/refresh_project.py
    arguments: None
    cpu: 2
    memory: 4
    short_summary: Create job to allow refreshing the project from source and rebuilding.
    environment:
      TASK_TYPE: CREATE/RUN_JOB

  - type: run_job
    entity_label: refresh_project
    short_summary: Run job to refresh the project from source and rebuilding.

  - type: start_application
    name: RagStudio
    subdomain: ragstudio
    bypass_authentication: false
    static_subdomain: false
    script: scripts/startup_app.py
    short_summary: Create and start RagStudio.
    long_summary: Create and start RagStudio.
    cpu: 2
    memory: 4
    environment_variables:
      TASK_TYPE: START_APPLICATION
      DB_URL: "jdbc:h2:file:~/databases/rag"
      LLM_SERVICE_URL: "http://localhost:8081"
      API_URL: "http://localhost:8080"