Giter Club home page Giter Club logo

xai-demonstrator / xai-demonstrator Goto Github PK

View Code? Open in Web Editor NEW
19.0 3.0 7.0 15.69 MB

The XAI Demonstrator is a modular platform that lets users interact with production-grade Explainable AI (XAI) systems.

Home Page: https://www.erklaerbare-ki.de/xai-demonstrator/

License: Apache License 2.0

Python 58.91% Dockerfile 1.58% HTML 2.97% JavaScript 12.71% Vue 22.29% Shell 0.76% CSS 0.78%
xai ai explainable-ai mobile-first pytorch huggingface-transformers fastapi vuejs machine-learning keras tensorflow docker

xai-demonstrator's Introduction

XAI Demonstrator

License Language grade: JavaScript Language grade: Python

The XAI Demonstrator is a modular platform that lets users interact with production-grade Explainable AI (XAI) systems.

Getting Started

If you just want to try the XAI Demonstrator, have a look at the list of publicly accessible deployments below.

Running the XAI Demonstrator locally requires Docker (but nothing else). If you're new to Docker, see the Docker Desktop installation instructions.

The fastest way to start a local version of the most recent development version of the XAI Demonstrator is to use the test-local deployment configuration:

cd deployment/test-local
docker-compose up

Then, you can reach the app at localhost:8000.

To start a local instance of the XAI Demonstrator built from source, run docker-compose up in the top-level directory, wait for all builds to complete, and access the app at localhost:8000.

Documentation

Documentation

A documentation with user guides is taking shape at xai-demonstrator.github.io/xai-demonstrator/.

Use Cases

Each use case illustrates a particular application of user-centric XAI methods.

Code Description Build Test
review-sentiment Explain the sentiment analysis of customer reviews by a multi-lingual BERT model using gradient-based feature importance explanation methods Review Sentiment BE: Coverage Status
FE: Coverage Status
visual-inspection Visually explain the classification of images using model-agnostic local explanation methods Visual Inspection BE: Coverage Status
FE: Coverage Status
guess-the-country (work in progress) Visual Inspection BE: Coverage Status
FE: Coverage Status

You can try out a standalone version each use case locally by running docker-compose up in the respective directory. Then, the use case is available at localhost:8000.

Experiment Tracking

It is possible to record user requests and additional data when the XAI Demonstrator is used in academic research or for user testing. To learn more, have a look at the experiment-tracker.

Package Description Build Test
experiment-proxy A proxy that records requests and responses to use cases for later analysis Experiment Tracker Coverage Status
experiment-proxy The service that collects and permanently stores the recorded data Experiment Tracker Coverage Status

Further Packages and Components

Package Description Build Test
common/backend-utils Python package providing common functionality for FastAPI backend services Backend Utils Coverage Status
common/vue-components NPM package providing common VueJS components for frontends Vue Components Coverage Status
landing-page Frontend that serves as a common entry point to all use cases. Landing Page

Deployments

Monitor Deployment(s)

Configuration Location Description Status
test-deployment https://test.xaidemo.de Latest version of the landing page and all use cases Test Deployment
prod-deployment https://www.xaidemo.de Current release Prod Deployment

License

The XAI Demonstrator is licensed under the terms of the Apache 2.0 license.

xai-demonstrator's People

Contributors

bergschaf avatar ddennyd avatar dependabot[bot] avatar ionicsolutions avatar larsmoe avatar matthias271828 avatar miker2241 avatar pns1123 avatar theresa323 avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar

xai-demonstrator's Issues

[BE] Alle Dependencies updaten

Die Backend-Dependencies vom Visual-Inspection-Use-Case auf den neuesten Stand bringen (vgl. die zahlreichen Dependabot-PRs, dort kannst du die aktuellen Versionen sehen)

[FE] Bild konfigurabel machen

Als Betreiber des Use Cases
möchte ich die gezeigte Bilddatei konfigurieren können
um auch ein anderes Bild anzuzeigen

[FE/BE] Konfidenz der KI anzeigen

Als Betreiber des Use Cases
möchte ich festlegen können, ob die Konfidenz der KI angezeigt werden soll
um Nutzer:innen darüber zu informieren

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.