Kyma Showcase

Overview

Kyma Showcase is a project about object recognition relying on GCP API and Serverless Functions Workflow. A customer uploads an image on the front page. Then the image is displayed in the feed. Clicking on any image in the feed redirects you to its details page showing all the information acquired from processing that image such as prevalent mood, labels describing the image itself, or objects in it. See the Details and Architecture sections for more information about Kyma Showcase.

Prerequisites

• Kyma cluster
• kubectl

Installation

To run Kyma Showcase application, follow these steps:

1. Clone the repository:

git clone https://github.com/kyma-incubator/kyma-showcase.git

2. Replace the placeholder values with the name of your cluster domain in the following files:

• resources/backend/apirule.yaml
• resources/frontend/apirule.yaml
• resources/frontend/frontend-config.yaml

3. Set Google Cloud Platform maps API key in the resources/frontend/frontend-config.yaml file.

4. Set secrets by replacing the placeholder values in the resources/secrets.yaml file.

5. Deploy the application on a Kubernetes cluster. Run the following command:

kubectl apply -f resources -R

6. Get the application URL. Run:

kubectl get apirule frontend -o jsonpath='{.spec.service.host}'

Development

To learn more, read the Development Guide.