Learning Objectives
- ELI5
- Skater
- What-If Tool
- Azure ML
Till now, we’ve learned about 2 Explainable AI Frameworks - LIME and SHAP.
There are more such tools/frameworks that can be utilized for model interpretation. Let’s have a look at those briefly. If you find any of them interesting, you can explore them in detail.
ELI5
- ELI5 is a Python package which helps debug machine learning classifiers and explain their predictions in an easy to understand and intuitive way.
- It is perhaps the easiest of the three machine learning frameworks to get started with since it involves minimal reading of documentation!
- However it doesn’t support true model-agnostic interpretations and support for models are mostly limited to tree-based and other parametric\linear models.
- You can install it using .
1pip install eli5
Skater
- Skater is a unified framework to enable Model Interpretation for all forms of models to help one build an Interpretable machine learning system often needed for real world use-cases using a model-agnostic approach.
- It is an open source python library designed to demystify the learned structures of a black box model both globally(inference on the basis of a complete data set) and locally(inference about an individual prediction).
- You can typically install Skater using a simple .
1pip install skater
What-If Tool
- The What-If Tool (WIT) provides an easy-to-use interface for expanding understanding of a black-box classification or regression ML model.
- With the plugin, you can perform inference on a large set of examples and immediately visualize the results in a variety of ways.
- The purpose of the tool is to give people a simple, intuitive, and powerful way to play with a trained ML model on a set of data through a visual interface with absolutely no code required.
- The tool can be accessed through TensorBoard or as an extension in a Jupyter or Colab notebook.
- A custom prediction function can be used to load any model, and provide additional customizations to the What-If Tool, including feature attribution methods like SHAP, Integrated Gradients, or SmoothGrad.
- Here’s the What-if Tool demo for the same dataset we used for SHAP implementation(UCI Dataset).
- Look at the beautiful interactive visualisations it creates: https://pair-code.github.io/what-if-tool/demos/uci.html
Azure ML
Azure ML proposes its own version of SHAP and alternative tools, adding interactive dashboards.
You can learn how to use its interpretability package here: https://docs.microsoft.com/en-us/azure/machine-learning/how-to-machine-learning-interpretability-aml
Additional Resources
This article outlines 8 Explainable AI Frameworks. You can go through it if you wish to explore more!