This is the pytorch implementation of the paper - Axiomatic Attribution for Deep Networks.

Last update: Dec 23, 2022

Overview

Integrated Gradients

This is the pytorch implementation of "Axiomatic Attribution for Deep Networks". The original tensorflow version could be found here.

Acknowledgement

ankurtaly's tensorflow version

Requirements

python-3.5.2
pytorch-0.4.1
opencv-python

TODO List

add more functions as the original code.
finetune the results, make them close to the original paper.

Instructions

Highly recommend to use GPU to accelerate the computation. If you use CPU, I will recommend to select some small networks, such as resnet18. You also need to put your images under examples/.

Lists of networks that support (of course, you can add any networks by yourself)

inception
resnet18
resnet152
vgg19

Run the code

python main.py --cuda --model-type='inception' --img='01.jpg'

Results

Results are slightly different from the original paper, it may have some bugs or need to do some adjustments. I will keep updating it, any contributions are welcome!

This is the pytorch implementation of the paper - Axiomatic Attribution for Deep Networks.

Related tags

Overview

Integrated Gradients

Acknowledgement

Requirements

TODO List

Instructions

Lists of networks that support (of course, you can add any networks by yourself)

Run the code

Results

Inception-v3

ResNet-18

ResNet-152

VGG-19

Owner

Tianhong Dai

Populating 3D Scenes by Learning Human-Scene Interaction https://posa.is.tue.mpg.de/

A Python-based development platform for automated trading systems - from backtesting to optimisation to livetrading.

codebase for "A Theory of the Inductive Bias and Generalization of Kernel Regression and Wide Neural Networks"

Distributional Sliced-Wasserstein distance code

code for paper "Does Unsupervised Architecture Representation Learning Help Neural Architecture Search?"

CCNet: Criss-Cross Attention for Semantic Segmentation (TPAMI 2020 & ICCV 2019).

Homepage of paper: Paint Transformer: Feed Forward Neural Painting with Stroke Prediction, ICCV 2021.

Research code for the paper "How Good is Your Tokenizer? On the Monolingual Performance of Multilingual Language Models"

Learning Multiresolution Matrix Factorization and its Wavelet Networks on Graphs

INSPIRED: A Transparent Dialogue Dataset for Interactive Semantic Parsing

MediaPipe is a an open-source framework from Google for building multimodal

Minimal But Practical Image Classifier Pipline Using Pytorch, Finetune on ResNet18, Got 99% Accuracy on Own Small Datasets.

Source code for the Paper: CombOptNet: Fit the Right NP-Hard Problem by Learning Integer Programming Constraints}

An end-to-end project on customer segmentation

Using knowledge-informed machine learning on the PRONOSTIA (FEMTO) and IMS bearing data sets. Predict remaining-useful-life (RUL).

SegNet model implemented using keras framework

Logsig-RNN: a novel network for robust and efficient skeleton-based action recognition

ICLR21 Tent: Fully Test-Time Adaptation by Entropy Minimization

This is an official implementation for "SimMIM: A Simple Framework for Masked Image Modeling".

Weakly-Supervised Semantic Segmentation Network with Deep Seeded Region Growing (CVPR 2018).