Code implementation of "Sparsity Probe: Analysis tool for Deep Learning Models"

Overview

Sparsity Probe: Analysis tool for Deep Learning Models

GitHub license made-with-python made-with-pytorch

This repository is a limited implementation of Sparsity Probe: Analysis tool for Deep Learning Models by I. Ben-Shaul and S. Dekel (2021).

Folded Ball Example

Downloading the Repo

git clone https://github.com/idobenshaul10/SparsityProbe.git
pip install -r requirements.txt

Requirements

torch==1.7.0
umap_learn==0.4.6
matplotlib==3.3.2
tqdm==4.49.0
seaborn==0.11.0
torchvision==0.8.1
numpy==1.19.2
scikit_learn==0.24.2
umap==0.1.1

Usage

The first step of using this Repo should be to look at this example: CIFAR10 Example. In this example, we demonstrate running the Sparsity-Probe on a trained Resnet18 on the CIFAR10 dataset, at selected layers.

Creating a new enviorment:

Create a new environment in the environments directory, inheriting from BaseEnviorment. This enviorment should include the train and test datasets(including the matching transforms), the model layers we want to test the alpha-scores on(see cifar10_env example), and the trained model.

Training a model:

It is possible to train a basic model with the train.py script, which uses an environment to load the model and the datasets. Example Usage: python train/train_mnist.py --output_path "results" --batch_size 32 --epochs 100

Running the Sparsity Probe

Done using the DL_smoothness.py script. Arguments:
trees - Number of trees in the forest.
depth - Maximum depth of each tree.
batch_size - batch used in the forward pass(when computing the layer outputs)
env_name - enviorment which is loaded to measure alpha-scores on
epsilon_1 - the epsilon_low used for the numerical approximation. By default, epsilon_high is inited as 4*epsilon_low
only_umap - only create umaps of the intermediate layers(without computing alpha-scores)
use_clustering - run KMeans on intermediate layers
calc_test - calculate test accuracy(More metrics coming soon)
output_folder - location where all outputs are saved
feature_dimension - to reduce computation costs, we compute the alpha-scores on the features after a dimensionality reduction technique has been applied. As of now, if the dim(layer_outputs)>feature_dimension, the TruncatedSVD is used to reduce dim(layer_outputs) to feature_dimension. Default feature_dimension is 2500.

Plotting Results

Result plots can be created using this script.

UMAP example

Acknowledgements

Our pretrained CIFAR10 Resnet18 network used in the example is taken from This Repo.

License

This repository is MIT licensed, as found in the LICENSE file.

Python implementation of ADD: Frequency Attention and Multi-View based Knowledge Distillation to Detect Low-Quality Compressed Deepfake Images, AAAI2022.

ADD: Frequency Attention and Multi-View based Knowledge Distillation to Detect Low-Quality Compressed Deepfake Images Binh M. Le & Simon S. Woo, "ADD:

2 Oct 24, 2022
Human Activity Recognition example using TensorFlow on smartphone sensors dataset and an LSTM RNN. Classifying the type of movement amongst six activity categories - Guillaume Chevalier

LSTMs for Human Activity Recognition Human Activity Recognition (HAR) using smartphones dataset and an LSTM RNN. Classifying the type of movement amon

Guillaume Chevalier 3.1k Dec 30, 2022
Code of U2Fusion: a unified unsupervised image fusion network for multiple image fusion tasks, including multi-modal, multi-exposure and multi-focus image fusion.

U2Fusion Code of U2Fusion: a unified unsupervised image fusion network for multiple image fusion tasks, including multi-modal (VIS-IR, medical), multi

Han Xu 129 Dec 11, 2022
Sequence-tagging using deep learning

Classification using Deep Learning Requirements PyTorch version = 1.9.1+cu111 Python version = 3.8.10 PyTorch-Lightning version = 1.4.9 Huggingface

Vineet Kumar 2 Dec 20, 2022
NVIDIA container runtime

nvidia-container-runtime A modified version of runc adding a custom pre-start hook to all containers. If environment variable NVIDIA_VISIBLE_DEVICES i

NVIDIA Corporation 938 Jan 06, 2023
这是一个facenet-pytorch的库,可以用于训练自己的人脸识别模型。

Facenet:人脸识别模型在Pytorch当中的实现 目录 性能情况 Performance 所需环境 Environment 注意事项 Attention 文件下载 Download 预测步骤 How2predict 训练步骤 How2train 参考资料 Reference 性能情况 训练数据

Bubbliiiing 210 Jan 06, 2023
Modeling Temporal Concept Receptive Field Dynamically for Untrimmed Video Analysis

Modeling Temporal Concept Receptive Field Dynamically for Untrimmed Video Analysis This is a PyTorch implementation of the model described in our pape

qzhb 6 Jul 08, 2021
Datasets, Transforms and Models specific to Computer Vision

vision Datasets, Transforms and Models specific to Computer Vision Installation First install the nightly version of OneFlow python3 -m pip install on

OneFlow 68 Dec 07, 2022
A TensorFlow implementation of SOFA, the Simulator for OFfline LeArning and evaluation.

SOFA This repository is the implementation of SOFA, the Simulator for OFfline leArning and evaluation. Keeping Dataset Biases out of the Simulation: A

22 Nov 23, 2022
Fit Fast, Explain Fast

FastExplain Fit Fast, Explain Fast Installing pip install fast-explain About FastExplain FastExplain provides an out-of-the-box tool for analysts to

8 Dec 15, 2022
Compressed Video Action Recognition

Compressed Video Action Recognition Chao-Yuan Wu, Manzil Zaheer, Hexiang Hu, R. Manmatha, Alexander J. Smola, Philipp Krähenbühl. In CVPR, 2018. [Proj

Chao-Yuan Wu 479 Dec 26, 2022
A complete end-to-end demonstration in which we collect training data in Unity and use that data to train a deep neural network to predict the pose of a cube. This model is then deployed in a simulated robotic pick-and-place task.

Object Pose Estimation Demo This tutorial will go through the steps necessary to perform pose estimation with a UR3 robotic arm in Unity. You’ll gain

Unity Technologies 187 Dec 24, 2022
Dynamic Neural Representational Decoders for High-Resolution Semantic Segmentation

Dynamic Neural Representational Decoders for High-Resolution Semantic Segmentation Requirements This repository needs mmsegmentation Training To train

Adelaide Intelligent Machines (AIM) Group 7 Sep 12, 2022
A web-based application for quick, scalable, and automated hyperparameter tuning and stacked ensembling in Python.

Xcessiv Xcessiv is a tool to help you create the biggest, craziest, and most excessive stacked ensembles you can think of. Stacked ensembles are simpl

Reiichiro Nakano 1.3k Nov 17, 2022
Fast, accurate and reliable software for algebraic CT reconstruction

KCT CBCT Fast, accurate and reliable software for algebraic CT reconstruction. This set of software tools includes OpenCL implementation of modern CT

Vojtěch Kulvait 4 Dec 14, 2022
Spontaneous Facial Micro Expression Recognition using 3D Spatio-Temporal Convolutional Neural Networks

Spontaneous Facial Micro Expression Recognition using 3D Spatio-Temporal Convolutional Neural Networks Abstract Facial expression recognition in video

Bogireddy Sai Prasanna Teja Reddy 103 Dec 29, 2022
An Api for Emotion recognition.

PLAYEMO Playemo was built from the ground-up with Flask, a python tool that makes it easy for developers to build APIs. Use Cases Is Python your langu

greek geek 2 Jul 16, 2022
[ACMMM 2021 Oral] Enhanced Invertible Encoding for Learned Image Compression

InvCompress Official Pytorch Implementation for "Enhanced Invertible Encoding for Learned Image Compression", ACMMM 2021 (Oral) Figure: Our framework

96 Nov 30, 2022
Codes for AAAI22 paper "Learning to Solve Travelling Salesman Problem with Hardness-Adaptive Curriculum"

Paper For more details, please see our paper Learning to Solve Travelling Salesman Problem with Hardness-Adaptive Curriculum which has been accepted a

14 Sep 30, 2022
StyleMapGAN - Official PyTorch Implementation

StyleMapGAN - Official PyTorch Implementation StyleMapGAN: Exploiting Spatial Dimensions of Latent in GAN for Real-time Image Editing Hyunsu Kim, Yunj

NAVER AI 425 Dec 23, 2022