Open-source alternatives to CapsNet Visualization

Yellowbrick is a machine learning visualization library and model diagnostic tool designed to analyze feature importance, target distributions, and model error metrics. It serves as a visual toolkit for diagnosing underfitting and overfitting through the use of validation and learning curves. The project provides specialized suites for evaluating predictive models and unsupervised learning. It enables the determination of optimal cluster counts via elbow methods and silhouette coefficients, and assesses classifier and regressor quality through ROC curves, confusion matrices, and residual plot

Bias Auditing & Fair ML Toolkit

Interpretability and explainability of data and machine learning models

Interpret is an interpretable machine learning library and glassbox model framework. It provides toolkits for training inherently transparent models and applying post-hoc explanation techniques to make machine learning predictions human-understandable. The framework distinguishes itself by integrating differential privacy into the training of interpretable models to prevent sensitive data from leaking through explanations. It also features a visualization tool for rendering interactive decision paths and model behavior. The library covers model explainability through feature importance calcu

Code for replicating the experiments in the paper Learning to Explain: An Information-Theoretic Perspective on Model Interpretation at ICML 2018, by Jianbo Chen, Mitchell Stern, Martin J. Wainwright, Michael I. Jordan.

Auralisation of learned features in CNN (for audio)

This project is a machine learning experiment tracker and event file generator that enables the recording of scalars, images, and histograms to monitor model performance. It functions as an integration bridge that allows training metrics from PyTorch to be logged into files compatible with the TensorBoard dashboard. The system includes a remote log synchronizer designed to stream experiment data to cloud services. This allows for the remote management and analysis of training results and the comparison of datasets across different training runs. The utility covers a broad range of monitoring

Netron is a visualizer for neural network and machine learning models. It provides a graphical interface that renders model architectures as interactive node-link diagrams, allowing users to inspect internal layers, tensors, and metadata. By performing static analysis, the tool enables the examination of model definitions without executing the underlying machine learning code. The software distinguishes itself through a schema-driven parsing engine that translates diverse proprietary model formats into a unified internal graph structure. This approach ensures interoperability, allowing users

Contrastive Explanation (Foil Trees), developed at TNO/Utrecht University

Code for "High-Precision Model-Agnostic Explanations" paper

This project is an agnostic model interpretability framework and explainability tool designed to provide local interpretable explanations for individual predictions. It functions as a local surrogate model that approximates the behavior of any machine learning classifier or regression model to identify the most influential features for a specific instance. The framework is designed to be model-agnostic, meaning it can explain predictions across tabular, text, and image data regardless of the underlying architecture. It employs local linear approximations and feature importance visualization t

moDel Agnostic Language for Exploration and eXplanation

An intuitive library to add plotting functionality to scikit-learn objects.

python partial dependence plot toolbox

Algorithms for explaining machine learning models

SHAP is a machine learning explainer that uses a game-theoretic framework to estimate the contribution of each feature to a model prediction. It provides a set of tools for quantifying how individual input features push a specific output away from a baseline value. The project includes specialized explainers for different architectures, including high-speed implementations for decision trees and ensemble models, linearization algorithms for deep learning networks, and covariance integration for linear models. It also features a model-agnostic interpretability tool that uses a kernel method to

A library for debugging/inspecting machine learning classifiers and explaining their predictions

Lucid is a TensorFlow interpretability toolkit and visualization library designed to analyze the internal representations of neural networks. It functions as a gradient-based optimization framework that generates images and atlases to reveal the features learned by specific neurons and layers. The library enables the creation of activation atlases and the mapping of high-dimensional neural activations into lower-dimensional spaces to study model behavior. It utilizes differentiable image parametrization to optimize visual inputs that maximally activate network components. The system covers a

Model analysis tools for TensorFlow