nxpsmcd/eiq-ai-toolkit-api repository overview

⁠Quick Reference

• Maintained by: NXP⁠
• Where to get help: NXP Community⁠
• Training materials: eIQ Training Materials⁠

⁠What is eIQ® AI Toolkit?

eIQ AI Toolkit is an advanced tool offering graphical and REST API interfaces for two primary purposes:

• Optimizing, converting and deploying neural network models through structured workflows
• Profiling neural network models

These capabilities enable users to fine-tune their models using a wide selection of algorithms, identify and resolve deployment issues, validate results, and measure model performance.

⁠How to Use This Image

This specific image contains only a single component of the eIQ AI Toolkit software stack. It is unusable on its own and requires additional docker images to be ran alongside it.

Therefore, it is recommended to run this image solely through the following docker-compose.yaml example:

services:
  # Runs database migrations for version upgrades with existing data. Skip for new deployments.
  migrate:
    image: nxpsmcd/eiq-ai-toolkit-api:latest
    entrypoint: []
    command: >
      python3 -m pymongo_migrate.cli upgrade
      -u mongodb://root:example@mongo:27017
      -d datastore
      -m /app/db-migration/pymongo_migrations
    depends_on:
      - mongo

  api:
    image: nxpsmcd/eiq-ai-toolkit-api:latest
    container_name: eiq-ai-toolkit-api
    restart: unless-stopped
    ports:
      - 8000:8000
    environment:
      - DATABASE_URL=mongodb://root:example@mongo:27017
      - REDIS_URL=redis://redis:6379
      - DOWNLOAD_SIGNING_SECRET=CHANGE_ME
    volumes:
      - app_data:/app_data

  workers:
    image: nxpsmcd/eiq-ai-toolkit-workers:latest
    container_name: eiq-ai-toolkit-workers
    restart: unless-stopped
    environment:
      - DATABASE_URL=mongodb://root:example@mongo:27017
      - REDIS_URL=redis://redis:6379
    volumes:
      - app_data:/app_data

  frontend:
    image: nxpsmcd/eiq-ai-toolkit-gui:latest
    container_name: eiq-ai-toolkit-gui
    restart: unless-stopped
    ports:
      - 8080:8080

  mongo:
    image: mongo
    container_name: eiq-ai-toolkit-mongo
    restart: unless-stopped
    ports:
      - 27017:27017
    environment:
      MONGO_INITDB_ROOT_USERNAME: root
      MONGO_INITDB_ROOT_PASSWORD: example
    volumes:
      - mongodata:/data/db
      - mongoconfig:/data/configdb

  redis:
    image: redis:alpine
    container_name: eiq-ai-toolkit-redis
    restart: unless-stopped
    ports:
      - 6379:6379
    command: redis-server --loglevel warning
    volumes:
      - redis_data:/data

volumes:
  app_data:
  mongodata:
  mongoconfig:
  redis_data:

If you want to use a specific version of eIQ AI Toolkit, you may change the image tag from latest to the desired version.

But please note that each eIQ AI Toolkit Docker image, like eiq-ai-toolkit-api, eiq-ai-toolkit-workers and eiq-ai-toolkit-gui needs to use the same version, otherwise compatibility cannot be guaranteed.

The Redis port and MongoDB credentials may be customized and propagated to eIQ AI Toolkit images using the DATABASE_URL and REDIS_URL environmental variables.

A variable DOWNLOAD_SIGNING_SECRET used for signing file download URLs should be set as well. Although, for purely local deployment, it should not be a problem to leave it as is.

After a successful launch of all the services, the graphical user interface can be accessed in the browser on URL localhost:8080 (or any other custom port if the frontend service port mapping was changed in the example).

Please, refer to the eIQ Training Materials⁠ for more detailed installation and deployment instructions.

⁠License

View license information⁠ for the software contained in this image.

As with all Docker images, these likely also contain other software which may be under other licenses (such as Bash, etc. from the base distribution, along with any direct or indirect dependencies of the primary software being contained).

As for any pre-built image usage, it is the image user's responsibility to ensure that any use of this image complies with any relevant licenses for all software contained within.