tcav
Concept explanation tool
An interpretability method that provides explanations for neural network predictions by highlighting high-level concepts relevant to classification tasks.
Code for the TCAV ML interpretability project
632 stars
34 watching
150 forks
Language: Jupyter Notebook
last commit: 4 months ago interpretabilitymachine-learningtcav
Related projects:
| Repository | Description | Stars |
|---|---|---|
 sicara/tf-explain | A library providing interpretability methods for TensorFlow 2.x models | 1,018 |
 understandable-machine-intelligence-lab/quantus | An eXplainable AI toolkit for evaluating and interpreting neural network explanations in various deep learning frameworks. | 556 |
| tensorflow/text | Preprocessing and processing tools for text data in machine learning models | 1,233 |
 infocusp/tf_cnnvis | A tool to visually analyze and understand deep learning models' internal workings, specifically convolutional neural networks. | 780 |
| tensorflow/model-analysis | Evaluates and visualizes the performance of machine learning models. | 1,258 |
 csinva/hierarchical-dnn-interpretations | Provides an implementation of Hierarchical explanations for Neural Network predictions | 125 |
 jbloomaus/decisiontransformerinterpretability | An open-source project that provides tools and utilities to understand how transformers are used in reinforcement learning tasks. | 73 |
 burness/tensorflow-101 | An end-to-end deep learning project demonstrating various TensorFlow techniques and applications in image classification, GANs, text classification, and more. | 1,113 |
 kwotsin/tensorflow-xception | An implementation of a deep learning model for computer vision tasks using TensorFlow | 207 |
 marcotcr/anchor | Provides a method to generate explanations for predictions made by any black box classifier. | 798 |
 trusted-ai/aix360 | A toolkit for explaining complex AI models and data-driven insights | 1,633 |
| tensorflow/mesh | A library that enables distributed deep learning by partitioning tensors across processors in a mesh topology. | 1,592 |
 interpretml/dice | Provides counterfactual explanations for machine learning models to facilitate interpretability and understanding. | 1,364 |
| tensorflow/haskell | Haskell bindings for a popular machine learning framework, allowing developers to build and deploy neural networks in the Haskell programming language. | 1,582 |
| tensorflow/tensorflow | An end-to-end platform for building and deploying machine learning applications | 186,382 |