Quantus

Visit Tool

Quantus is an eXplainable AI toolkit that provides responsible evaluation of neural network explanations. It offers over 35 metrics across 6 categories for XAI evaluation, supporting PyTorch and TensorFlow models.

Claim this tool

1View

At a glance

Pricing

Open Source

Free tier

Yes

API

Skill level

Technical

About

What is Quantus?

Quantus is an open-source eXplainable AI (XAI) toolkit designed for the responsible evaluation of neural network explanations. It offers a comprehensive suite of over 35 metrics categorized into faithfulness, robustness, localisation, complexity, randomisation (sensitivity), and axiomatic properties. The toolkit supports various data types including image, time-series, tabular, and NLP (upcoming), and is compatible with both PyTorch and TensorFlow models. Quantus aims to provide richer insights into how different XAI methods compare, moving beyond simple visual comparisons to offer holistic quantification and sensitivity analysis. It also includes built-in support for popular explanation methods like Captum, tf-explain, and Zennit, making it a versatile tool for researchers and developers in the AI field.

Best used for

Ideal for data scientists and developers who need to rigorously evaluate the quality of neural network explanations, compare different XAI methods quantitatively, and ensure the reliability of their AI models. Especially valuable for researchers working on interpretability and responsible AI development.

Common actions

evaluate AI explanations

compare XAI methods

quantify explanation quality

assess neural network interpretability

workflowscollaborationautomated workflowopen-sourcedeepfakelow-code/no-codegithub copilotface swapping"AI Agents"

Capabilities

Key features

35+ XAI evaluation metrics
PyTorch and TensorFlow support
Image, time-series, tabular data
Built-in explanation method support
Faithfulness metrics
Robustness metrics
Localisation metrics

Target Audience

data scientistdeveloperresearcher

Integrations

Not yet documented

Pricing & Plans

Open Source

Free

FAQs

What types of neural network explanations can Quantus evaluate?

Quantus can evaluate explanations from various neural network models, supporting both PyTorch and TensorFlow frameworks. It is designed to work with different data types, including image, time-series, and tabular data, with upcoming support for NLP explanations.

How many evaluation metrics does Quantus offer and what categories do they cover?

Quantus offers over 35 evaluation metrics categorized into six main areas: faithfulness, robustness, localisation, complexity, randomisation (sensitivity), and axiomatic properties. This comprehensive set allows for a multi-faceted assessment of XAI methods.

Does Quantus provide support for existing explanation methods?

Yes, Quantus includes extended built-in support for several popular explanation methods. These include Captum, tf-explain, and Zennit, making it easier to integrate and evaluate explanations generated by these tools.

Trending

Subcategories trending in AI Agents & Automation

Chatbots & Conversational AI General-Purpose Agents Workflow Agents Personal Assistants RAG & Document AI Voice Agents

Trending

Explore

Browse AI tools by category

Content & Design Productivity & Business Coding & Development AI Agents & Automation Research & Education Wellness & Lifestyle Career Development Marketing & Growth Data & Analytics Customer Support & CX Finance E-commerce