OMMR4all-deploy

Deployment/Setup of all ommr4all services

Deployment-Setup with Docker

You can setup OMMR4all using docker and docker-compose.

Initial setup

1. Download and install docker-ce and docker-compose for your platform.
2. Download the docker-compose.yml file.
3. Open docker-compose.yml and replace ${STORAGE} and the ${PORT} to your wishes (e.g., use /opt/ommr4all-storage and 8001).
4. Build the container and bring it up:

docker-compose up -d

5. Create a super user:

docker-compose run /opt/ommr4all/ommr4all-deploy-venv/bin/python /opt/ommr4all/ommr4all-deploy/modules/ommr4all-server/manage.py createsuperuser

Updating

1. docker-compose pull
2. docker-compuse up

You can run docker image prune -f to clean all previous versions or older images that are currently unused.

Deployment-Setup without docker

Follow the instructions in the Dockerfile. You can also setup a gitlab-runner for automatic deployment (Clone the project on github.com with CI-integration), create a runner with either

• deployment-production: redeploy if a new (version) tag was added
• deployment-master: redeploy if the master is updated

Development-Setup (Any operating system)

These instructions are not complete yet.

1. Download and install all requirements (node>=10, >=python3.6)
2. Install the IDEs (IntelliJ, or PyCharm and WebStorm)
3. Create a virtual environment, activate it, and install your desired tensorflow version (e.g., pip install tensorflow_gpu<2)
4. Install all python submodules (located in the modules directure) but the server: python setup.py install.
5. Install the server requirements.txt: pip install -r requirements.txt.
6. Open the ommr4all-client directory in WebStorm and launch the Angluar CLI Server.
7. Open the ommr4all-sever directory in PyCharm and launch the Django Server.
8. In WebStorm launch the Angular Application which will open a browser.