model-interpretability

pytorch实现Grad-CAM和Grad-CAM++,可以可视化任意分类网络的Class Activation Map (CAM)图,包括自定义的网络;同时也实现了目标检测faster r-cnn和retinanet两个网络的CAM图;欢迎试用、关注并反馈问题...

Pytorch Implementation of recent visual attribution methods for model interpretability

Interesting resources related to Explainable Artificial Intelligence, Interpretable Machine Learning, Interactive Machine Learning, Human in Loop and Visual Analytics.

Class Activation Map (CAM) Visualizations in PyTorch.

A set of tools for leveraging pre-trained embeddings, active learning and model explainability for effecient document classification

Overview of different model interpretability libraries.

Official repository for the paper "Instance-wise Causal Feature Selection for Model Interpretation" (CVPRW 2021)

Used the Functional API to built custom layers and non-sequential model types in TensorFlow, performed object detection, image segmentation, and interpretation of convolutions. Used generative deep learning including Auto Encoding, VAEs, and GANs to create new content.

Human Factors in Model Interpretability: Industry Practices, Challenges, and Needs

Code for "Investigating and Simplifying Masking-based Saliency Methods for Model Interpretability" (https://arxiv.org/abs/2010.09750)

AI to Predict Yield in Aeroponics

A python script for basic data cleaning/manipulation and modelling based on the open source House Sales Advanced Regression Techniques(Kaggle)

To predict the rating of a developer using various data captured during an online test

A major gas and electricity utility that supplies to SME. The power-liberalization of the energy market in Europe has led to significant customer churn.Building a churn model to understand whether price sensitivity is the largest driver of churn.Verifying the hypothesis of price sensitivity being to some extent correlated with churn.

Visualizing an XGBoost model in R using a sunburst plot (using inTrees)

A machine learning project developing classification models to predict COVID-19 diagnosis in paediatric patients.

Model interpretability for Explainable Artificial Intelligence

Implementation of the Grad-CAM algorithm in an easy-to-use class, optimized for transfer learning projects and written using Keras and Tensorflow 2.x

Course project for 6.869: automatic summarization for neural net interpretability

MSc dissertation project, written in using LaTeX.

Using LIME and SHAP for model interpretability of Machine Learning Black-box models.

Interpretability and Fairness in Machine Learning

The project provides explanation of what SHAP is and how it can be used to interpret model. Also contains Notebook with detail on model interpretability method SHAP and code implementation on Heart disease dataset.

Will They Pay? A machine learning solution to understand mobile app user payment behavior

The "keras-translator" helps you to understand a keras trained model.

Code for "High-Precision Model-Agnostic Explanations" paper. A follow up to LIME model.

Using machine learning models to predict if patients have chronic kidney disease based on a few features. The results of the models are also interpreted to make it more understandable to health practitioners.

erformed a predictive analysis on the customer's Bank Loan Application data to predict loan status. Using python, pandas, scipy, seaborn, AutoML libraries, and machine learning techniques. Used Machine Learning techniques to accurately predict the evaluation scheme if the particular loan will be 'Fully Paid' or 'Charged Off'. This means if Bank accepts a particular person's loan application will it be 'Fully Paid' or 'Charged Off'