README.md

Open-Source CI/CD platform for ML teams

Giskard creates interfaces for humans to inspect AI models

Documentation • Getting Started • Upload Model & Data • Blog • Discord Community • Advisors

About Giskard

Giskard creates interfaces for humans to inspect AI models. It is open-source & self-hosted.

⚡ Explore and Validate - Collaborate with business stakeholders with direct feedback & discussion.
🧑‍🔧 Test - Exhaustive test suites, backed by State-of-the-Art ML research. 🧑‍🔧 Automate - Protect your ML models against the risk of regressions, drift and bias.

---

Interactive demo

Click the image below to start the demo:

Product workflow

Installation

Requirements: git, docker and docker-compose

git clone https://github.com/Giskard-AI/giskard.git
cd giskard
docker-compose up -d

After the application is started you can access at:

http://localhost:19000 with default login / password: admin / admin

Upload your model

Easy upload for any Python model: PyTorch, TensorFlow, Transformers, Scikit-learn, etc.
👉 Documentation

Collect feedback on your model

Improve ML models with business stakeholders in no time.
👉 Documentation

Get automated test suites in seconds

Exhaustive test suites, backed by 
State-of-the-Art ML research.
👉 Documentation

Deploy tests in your CI/CD Pipeline

Protect your ML models against the risk of regressions, drift and bias.
👉 Documentation

How to contribute

We welcome contributions from the Machine Learning community!

Read this guide to get started.