- CI/CD Pipeline for Flask ML Deploy
- Generate SSH keys and connect your Azure environment with your Github account.
- ssh-keygen -t rsa
- Clone the repo using Azure cloud, and git commands
- Create the python virtual environment and activate it my repo. Execute make all command to install all dependencies.
- python3 -m venv ~/.CI-CD-Pipeline source ~/.CI-CD-Pipeline/bin/activate
- Make a new project in Microsoft Azure, define organization and connect to Github repo.
- Create environment variable and activate it using python3.
- Run python app and test the make_prediction.sh script if it works
- Run the webapp with your app name (FlaskML2022) and check its log trails
- Verify the respective pipelines in Azure pipelines and check build and deployment jobs
-
In make_predict_azure_app.sh, change post X to the value of your own webapp that was created earlier.
-
Verify the make_predict_azure_app.sh again with your app URL
Video demonstration here