# Setting up CHAP Prediction App in DHIS2 This document describes how to install the Prediction App in your DHIS2 instance. **Requirements:** - DHIS2 version **2.39+** ## 1 Setting up CHAP-Core REST API 1. The Prediction App requires communication with an instance of the CHAP-Core REST-API. For using the Prediction App in a production environment, it's recommended that the Chap REST-API has been [set up on a permanent server by a system admin](running-chap-on-server). > 🛈 Alternatively, you can follow [the instructions on how to setup a Chap REST-API locally using Docker](../installation/docker-compose-doc.md). 2. Verify that you have access to the CHAP REST-API and that it's running correctly, by visiting `/docs`. A Swagger page (REST-API documentation) should display. > 🛈 If you're running the REST-API as a local Docker container, verify by going to `http://localhost:8000/docs`. ## 2 Download and Install Prediction App Currently the Prediction App is not yet available on the DHIS2 App Hub. Until then you need to download the Prediction App as a ZIP file for installing it on your DHIS2 instance. 1. Download Prediction App from the versions available at [https://github.com/dhis2/prediction-app/tree/refs/heads/main/build/bundle](https://github.com/dhis2/prediction-app/tree/refs/heads/main/build/bundle) 2. Upload the ZIP file (Prediction App) to your DHIS2 instance ([Learn how to install a custom DHIS2 app](https://dhis2-app-course.ifi.uio.no/learn/dhis2/app-development-guides/build-publish-app-platform-app/manual-install/)) > 🛈 Alternatively for developers, you can run a local development version of the Prediction App [as described here](https://github.com/dhis2/prediction-app/blob/main/doc/development/README.md). ## 3 Configure CHAP Core URL Finally, the Prediction App needs to know which URL to use to communicate with the CHAP Core REST-API. By clicking "Edit CHAP Core URL" located at the right side menu, a modal should display where you can configure this setting. Set this value to ``. > 🛈 If you're running the REST-API as a local Docker container, this value should be set to `http://localhost:8000`. drawing After clicking "Save", Prediction App should reload and start to fetch and send data to the provided URL.