Displaying a Disasters Image Classification Model Results from a Restful API

This is a supplementary submission of final paper for the CIS726 course.

It contains the code necessary to issue a RESTful API request by uploading an image, and classify it to a disaster type with XAI annotations.

To view a live example, click here.

NOTE: The environment that the current web application is deployed on cannot sustain multiple concurrent requests. Also, the request takes a considerable amount of time to process, allow up to 80 seconds.

Getting Started

Clone the project from GitHub

$ git clone https://github.com/tariqshaban/disaster-classification-with-xai-visualizer.git

The project is an implementation of disaster-classification-with-xai repository

This project was generated with Angular CLI version 13.2.6.

API setup

The project mandates a working API to retrieve the results, clone the following project from GitHub and setup a local server, Python must be installed since it operates on FastAPI. $ git clone https://github.com/tariqshaban/disaster-classification-with-xai-server.git

Configure api-endpoint.service.ts properties as necessary to identify the API endpoints.

Make sure to handle the CORS policy if the server is running locally.

Development server

Run ng serve for a dev server. Navigate to http://localhost:4200/. The app will automatically reload if you change any of the source files.

---