This is a simple iris flower classification model deployment project as flask app.
Directly run on Docker without building
docker run --rm -d -p 8080:8080 erkansirin78/flask-iris-classification:2021-3
Build an image (Dockerize) and run on Docker container
- Download project
git clone https://github.com/erkansirin78/flask-iris-classification.git
cd flask-iris-classification
-
Build image
docker image build -t my_flask_iris:1.0 . -
Run container
docker run --rm --name flask_iris -p 8080:8080 -d my_flask_iris:1.0 -
Open browser http://localhost:8080/
Enjoy your predictions.
On Kubernetes
-
You can deploy the model/app on Kubernetes with deployment object.
-
You can deploy this model/app on Openshift just using git project url with OpenShift build menu.
Input web ui:
Predictions:
