Official PyTorch implementation and pretrained models of the paper Self-Supervised Classification Network

Overview

Self-Classifier: Self-Supervised Classification Network

Official PyTorch implementation and pretrained models of the paper Self-Supervised Classification Network. Self-Classifier is a self-supervised end-to-end classification neural network. It learns labels and representations simultaneously in a single-stage end-to-end manner.

Self-Classifier architecture. Two augmented views of the same image are processed by a shared network. The cross-entropy of the two views is minimized to promote same class prediction while avoiding degenerate solutions by asserting a uniform prior. The resulting model learns representations and class labels in a single-stage end-to-end unsupervised manner. CNN: Convolutional Neural Network; FC: Fully Connected.

Setup

  1. Install Conda environment:

     conda env create -f ./environment.yml
    
  2. Install Apex with CUDA extension:

     export TORCH_CUDA_ARCH_LIST="7.0"  # see https://en.wikipedia.org/wiki/CUDA#GPUs_supported
     pip install git+git://github.com/NVIDIA/[email protected] --install-option="--cuda_ext"         
    

Training & Evaluation

Distributed training & evaluation is available via Slurm. See SBATCH scripts here.

IMPORTANT: set DATASET_PATH, EXPERIMENT_PATH and PRETRAINED_PATH to match your local paths.

Training

For training self-classifier on 4 nodes of 4 GPUs each for 800 epochs run:

    sbatch ./scripts/train.sh

Evaluation

Image Classification with Linear Models

For training a supervised linear classifier on a frozen backbone, run:

    sbatch ./scripts/eval.sh

Unsupervised Image Classification

For computing unsupervised image classification metrics (NMI: Normalized Mutual Information, AMI: Adjusted Normalized Mutual Information and ARI: Adjusted Rand-Index) and generating qualitative examples, run:

    sbatch ./scripts/cls_eval.sh

Image Classification with kNN

For running K-nearest neighbor classifier on ImageNet validation set, run:

    sbatch ./scripts/knn_eval.sh

Ablation study

For training the 100-epoch ablation study baseline, run:

    sbatch ./scripts/ablation/train_100ep.sh

For training any of the ablation study runs presented in the paper, run:

    sbatch ./scripts/ablation//.sh

Pretrained Models

Download pretrained 100/800 epochs models here.

Qualitative Examples (classes predicted by Self-Classifier on ImageNet validation set)

Low entropy classes predicted by Self-Classifier on ImageNet validation set. Images are sampled randomly from each predicted class. Note that the predicted classes capture a large variety of different backgrounds and viewpoints.

To reproduce qualitative examples, run:

    sbatch ./scripts/cls_eval.sh

License

See the LICENSE file for more details.

Citation

If you find this repository useful in your research, please cite:

@article{amrani2021self,
  title={Self-Supervised Classification Network},
  author={Amrani, Elad and Bronstein, Alex},
  journal={arXiv preprint arXiv:2103.10994},
  year={2021}
}
Owner
Elad Amrani
Machine Learning (EE) MSc Student at Technion
Elad Amrani
CATE: Computation-aware Neural Architecture Encoding with Transformers

CATE: Computation-aware Neural Architecture Encoding with Transformers Code for paper: CATE: Computation-aware Neural Architecture Encoding with Trans

16 Dec 27, 2022
pytorch implementation of ABC : Auxiliary Balanced Classifier for Class-imbalanced Semi-supervised Learning

ABC:Auxiliary Balanced Classifier for Class-imbalanced Semi-supervised Learning, NeurIPS 2021 pytorch implementation of ABC : Auxiliary Balanced Class

Hyuck Lee 25 Dec 22, 2022
Monocular Depth Estimation Using Laplacian Pyramid-Based Depth Residuals

LapDepth-release This repository is a Pytorch implementation of the paper "Monocular Depth Estimation Using Laplacian Pyramid-Based Depth Residuals" M

Minsoo Song 205 Dec 30, 2022
Randomizes the warps in a stock pokeemerald repo.

pokeemerald warp randomizer Randomizes the warps in a stock pokeemerald repo. Usage Instructions Install networkx and matplotlib via pip3 or similar.

Max Thomas 6 Mar 17, 2022
Library for 8-bit optimizers and quantization routines.

bitsandbytes Bitsandbytes is a lightweight wrapper around CUDA custom functions, in particular 8-bit optimizers and quantization functions. Paper -- V

Facebook Research 687 Jan 04, 2023
FS-Mol: A Few-Shot Learning Dataset of Molecules

FS-Mol is A Few-Shot Learning Dataset of Molecules, containing molecular compounds with measurements of activity against a variety of protein targets. The dataset is presented with a model evaluation

Microsoft 114 Dec 15, 2022
R3Det based on mmdet 2.19.0

R3Det: Refined Single-Stage Detector with Feature Refinement for Rotating Object Installation # install mmdetection first if you haven't installed it

SJTU-Thinklab-Det 38 Dec 15, 2022
MAterial del programa Misión TIC 2022

Mision TIC 2022 Esta iniciativa, aparece como respuesta frente a los retos de la Cuarta Revolución Industrial, y tiene como objetivo la formación de 1

6 May 25, 2022
Distance correlation and related E-statistics in Python

dcor dcor: distance correlation and related E-statistics in Python. E-statistics are functions of distances between statistical observations in metric

Carlos Ramos Carreño 108 Dec 27, 2022
HyperPose is a library for building high-performance custom pose estimation applications.

HyperPose is a library for building high-performance custom pose estimation applications.

TensorLayer Community 1.2k Jan 04, 2023
PRIME: A Few Primitives Can Boost Robustness to Common Corruptions

PRIME: A Few Primitives Can Boost Robustness to Common Corruptions This is the official repository of PRIME, the data agumentation method introduced i

Apostolos Modas 34 Oct 30, 2022
Pytorch implementation of few-shot semantic image synthesis

Few-shot Semantic Image Synthesis Using StyleGAN Prior Our method can synthesize photorealistic images from dense or sparse semantic annotations using

40 Sep 26, 2022
Transfer Learning Remote Sensing

Transfer_Learning_Remote_Sensing Simulation R codes for data generation and visualizations are in the folder simulation. Experiment: California Housin

2 Jun 21, 2022
Official repository for "On Generating Transferable Targeted Perturbations" (ICCV 2021)

On Generating Transferable Targeted Perturbations (ICCV'21) Muzammal Naseer, Salman Khan, Munawar Hayat, Fahad Shahbaz Khan, and Fatih Porikli Paper:

Muzammal Naseer 46 Nov 17, 2022
Dynamic Attentive Graph Learning for Image Restoration, ICCV2021 [PyTorch Code]

Dynamic Attentive Graph Learning for Image Restoration This repository is for GATIR introduced in the following paper: Chong Mou, Jian Zhang, Zhuoyuan

Jian Zhang 84 Dec 09, 2022
PyTorch Lightning + Hydra. A feature-rich template for rapid, scalable and reproducible ML experimentation with best practices. ⚡🔥⚡

Lightning-Hydra-Template A clean and scalable template to kickstart your deep learning project 🚀 ⚡ 🔥 Click on Use this template to initialize new re

Łukasz Zalewski 2.1k Jan 09, 2023
Unofficial PyTorch Implementation of "DOLG: Single-Stage Image Retrieval with Deep Orthogonal Fusion of Local and Global Features"

Pytorch Implementation of Deep Orthogonal Fusion of Local and Global Features (DOLG) This is the unofficial PyTorch Implementation of "DOLG: Single-St

DK 96 Jan 06, 2023
Image Segmentation using U-Net, U-Net with skip connections and M-Net architectures

Brain-Image-Segmentation Segmentation of brain tissues in MRI image has a number of applications in diagnosis, surgical planning, and treatment of bra

Angad Bajwa 8 Oct 27, 2022
This repo includes our code for evaluating and improving transferability in domain generalization (NeurIPS 2021)

Transferability for domain generalization This repo is for evaluating and improving transferability in domain generalization (NeurIPS 2021), based on

gordon 9 Nov 29, 2022
Earthquake detection via fiber optic cables using deep learning

Earthquake detection via fiber optic cables using deep learning Author: Fantine Huot Getting started Update the submodules After cloning the repositor

Fantine 4 Nov 30, 2022