search

conrad-blucher-institute / semaphore

Semaphore is a student-built and maintained Python application that semi-automates the process of operationalizing AI models. Semaphore downloads data, runs models, post-process outputs, and saves outputs.
Creative Commons Zero v1.0 Universal
1 stars 0 forks source link
ai2es

readme

Semaphore

Description

Semaphore is a student-built and maintained Python application that semi-automates the process of operationalizing AI models. Semaphore downloads data, runs models, post-process outputs, and saves outputs. Semaphore currently takes model data in the form of .H5 files and input specifications and a DSPEC JSON. Examples of both are included in this repository.

Dependency

Deployment

  1. You will need to create a .env file in the root directory of the project. This file will contain the environment variables for the docker container. You can copy the contents of env.dist and replace the values with your own.
  2. Ensure that Docker Desktop is running on your machine
  3. Run docker compose build and docker compose up (run docker compose up -d to run in the background)
  4. Intitalize the database by running: docker exec semaphore-core python3 tools/migrate_db.py
  5. Database can be accessed using pgAdmin or VSCode Postgresql Explorer plugin Register DB to pgAdmin Enter your database credentials to register to pgAdmin
  6. To stop the containers: Press ctrl+C in the terminal or if running in background close in Docker Desktop (docker compose down also works)
  7. If you are making changes to the code, be sure to docker compose down and docker compose build before docker compose up. If you are making changes to the database, be sure to delete the existing DB volume on Docker Desktop before building again

Running Semaphore

Currently, you can run Semaphore through the make_and_save_prediction and make_prediction methods in the src/semaphoreRunner.py which contains documentation for the CLI. Keep in mind the DB will need to be loaded with the proper location mappings for your ingested data. Feel free to reach out to one of the authors for help getting Semaphore up and running.

Running FastAPI Application

  1. Make sure your environment is activated! source ./venv/vsem/Scripts/activate
  2. pip install -r requirements.txt to install new dependencies (if you haven't already)
  3. From the root directory, run uvicorn src.API.apiDriver:app
  4. You can specify --host and --port if you so desire
  5. uvicorn will give you a link to the index page of the API. Add /docs to the URL to access the docs page

Authors