A Pytorch implementation of CVPR 2021 paper "RSG: A Simple but Effective Module for Learning Imbalanced Datasets"

Related tags

Deep LearningRSG
Overview

RSG: A Simple but Effective Module for Learning Imbalanced Datasets (CVPR 2021)

A Pytorch implementation of our CVPR 2021 paper "RSG: A Simple but Effective Module for Learning Imbalanced Datasets". RSG (Rare-class Sample Generator) is a flexible module that can generate rare-class samples during training and can be combined with any backbone network. RSG is only used in the training phase, so it will not bring additional burdens to the backbone network in the testing phase.

How to use RSG in your own networks

  1. Initialize RSG module:

    from RSG import *
    
    # n_center: The number of centers, e.g., 15.
    # feature_maps_shape: The shape of input feature maps (channel, width, height), e.g., [32, 16, 16].
    # num_classes: The number of classes, e.g., 10.
    # contrastive_module_dim: The dimention of the contrastive module, e.g., 256.
    # head_class_lists: The index of head classes, e.g., [0, 1, 2].
    # transfer_strength: Transfer strength, e.g., 1.0.
    # epoch_thresh: The epoch index when rare-class samples are generated: e.g., 159.
    
    self.RSG = RSG(n_center = 15, feature_maps_shape = [32, 16, 16], num_classes=10, contrastive_module_dim = 256, head_class_lists = [0, 1, 2], transfer_strength = 1.0, epoch_thresh = 159)
    
    
  2. Use RSG in the forward pass during training:

    out = self.layer2(out)
    
    # feature_maps: The input feature maps.
    # head_class_lists: The index of head classes.
    # target: The label of samples.
    # epoch: The current index of epoch.
    
    if phase_train == True:
      out, cesc_total, loss_mv_total, combine_target = self.RSG.forward(feature_maps = out, head_class_lists = [0, 1, 2], target = target, epoch = epoch)
     
    out = self.layer3(out) 
    

The two loss terms, namely ''cesc_total'' and ''loss_mv_total'', will be returned and combined with cross-entropy loss for backpropagation. More examples and details can be found in the models in the directory ''Imbalanced_Classification/models''.

How to train

Some examples:

Go into the "Imbalanced_Classification" directory.

  1. To reimplement the result of ResNet-32 on long-tailed CIFAR-10 ($\rho$ = 100) with RSG and LDAM-DRW:

    Export CUDA_VISIBLE_DEVICES=0,1
    python cifar_train.py --imb_type exp --imb_factor 0.01 --loss_type LDAM --train_rule DRW
    
  2. To reimplement the result of ResNet-32 on step CIFAR-10 ($\rho$ = 50) with RSG and Focal loss:

    Export CUDA_VISIBLE_DEVICES=0,1
    python cifar_train.py --imb_type step --imb_factor 0.02 --loss_type Focal --train_rule None
    
  3. To run experiments on iNaturalist 2018, Places-LT, or ImageNet-LT:

    Firstly, please prepare datasets and their corresponding list files. For the convenience, we provide the list files in Google Drive and Baidu Disk.

    Google Drive Baidu Disk
    download download (code: q3dk)

    To train the model:

    python inaturalist_train.py
    

    or

    python places_train.py
    

    or

    python imagenet_lt_train.py
    

    As for Places-LT or ImageNet-LT, the model is trained on the training set, and the best model on the validation set will be saved for testing. The "places_test.py" and 'imagenet_lt_test.py' are used for testing.

Citation

@inproceedings{Jianfeng2021RSG,
  title = {RSG: A Simple but Effective Module for Learning Imbalanced Datasets},
  author = {Jianfeng Wang and Thomas Lukasiewicz and Xiaolin Hu and Jianfei Cai and Zhenghua Xu},
  booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
  year={2021}
}
Sionna: An Open-Source Library for Next-Generation Physical Layer Research

Sionna: An Open-Source Library for Next-Generation Physical Layer Research Sionnaβ„’ is an open-source Python library for link-level simulations of digi

NVIDIA Research Projects 313 Dec 22, 2022
Bravia core script for python

Bravia-Core-Script You need to have a mandatory account If this L3 does not work, try another L3. enjoy

5 Dec 26, 2021
(ImageNet pretrained models) The official pytorch implemention of the TPAMI paper "Res2Net: A New Multi-scale Backbone Architecture"

Res2Net The official pytorch implemention of the paper "Res2Net: A New Multi-scale Backbone Architecture" Our paper is accepted by IEEE Transactions o

Res2Net Applications 928 Dec 29, 2022
NL-Augmenter 🦎 β†’ 🐍 A Collaborative Repository of Natural Language Transformations

NL-Augmenter 🦎 β†’ 🐍 The NL-Augmenter is a collaborative effort intended to add transformations of datasets dealing with natural language. Transformat

684 Jan 09, 2023
sssegmentation is a general framework for our research on strongly supervised semantic segmentation.

sssegmentation is a general framework for our research on strongly supervised semantic segmentation.

445 Jan 02, 2023
πŸ’‘ Type hints for Numpy

Type hints with dynamic checks for Numpy! (❒) Installation pip install nptyping (❒) Usage (❒) NDArray nptyping.NDArray lets you define the shape and

Ramon Hagenaars 377 Dec 28, 2022
code for our ECCV-2020 paper: Self-supervised Video Representation Learning by Pace Prediction

Video_Pace This repository contains the code for the following paper: Jiangliu Wang, Jianbo Jiao and Yunhui Liu, "Self-Supervised Video Representation

Jiangliu Wang 95 Dec 14, 2022
Training Very Deep Neural Networks Without Skip-Connections

DiracNets v2 update (January 2018): The code was updated for DiracNets-v2 in which we removed NCReLU by adding per-channel a and b multipliers without

Sergey Zagoruyko 585 Oct 12, 2022
Source code for the BMVC-2021 paper "SimReg: Regression as a Simple Yet Effective Tool for Self-supervised Knowledge Distillation".

SimReg: A Simple Regression Based Framework for Self-supervised Knowledge Distillation Source code for the paper "SimReg: Regression as a Simple Yet E

9 Oct 15, 2022
Official implementation for (Refine Myself by Teaching Myself : Feature Refinement via Self-Knowledge Distillation, CVPR-2021)

FRSKD Official implementation for Refine Myself by Teaching Myself : Feature Refinement via Self-Knowledge Distillation (CVPR-2021) Requirements Pytho

75 Dec 28, 2022
Reverse engineer your pytorch vision models, in style

πŸ” Rover Reverse engineer your CNNs, in style Rover will help you break down your CNN and visualize the features from within the model. No need to wri

Mayukh Deb 32 Sep 24, 2022
PuppetGAN - Cross-Domain Feature Disentanglement and Manipulation just got way better! πŸš€

Better Cross-Domain Feature Disentanglement and Manipulation with Improved PuppetGAN Quite cool... Right? Introduction This repo contains a TensorFlow

Giorgos Karantonis 5 Aug 25, 2022
The code for paper "Learning Implicit Fields for Generative Shape Modeling".

implicit-decoder The tensorflow code for paper "Learning Implicit Fields for Generative Shape Modeling", Zhiqin Chen, Hao (Richard) Zhang. Project pag

Zhiqin Chen 353 Dec 30, 2022
City-Scale Multi-Camera Vehicle Tracking Guided by Crossroad Zones Code

City-Scale Multi-Camera Vehicle Tracking Guided by Crossroad Zones Requirements Python 3.8 or later with all requirements.txt dependencies installed,

88 Dec 12, 2022
UniLM AI - Large-scale Self-supervised Pre-training across Tasks, Languages, and Modalities

Pre-trained (foundation) models across tasks (understanding, generation and translation), languages (100+ languages), and modalities (language, image, audio, vision + language, audio + language, etc.

Microsoft 7.6k Jan 01, 2023
project page for VinVL

VinVL: Revisiting Visual Representations in Vision-Language Models Updates 02/28/2021: Project page built. Introduction This repository is the project

308 Jan 09, 2023
YOLOv5 πŸš€ is a family of object detection architectures and models pretrained on the COCO dataset

YOLOv5 πŸš€ is a family of object detection architectures and models pretrained on the COCO dataset, and represents Ultralytics open-source research int

ι˜Ώζ‰ 73 Dec 16, 2022
Implementation of ViViT: A Video Vision Transformer

ViViT: A Video Vision Transformer Unofficial implementation of ViViT: A Video Vision Transformer. Notes: This is in WIP. Model 2 is implemented, Model

Rishikesh (ΰ€‹ΰ€·ΰ€Ώΰ€•ΰ₯‡ΰ€Ά) 297 Jan 06, 2023
Code base for the paper "Scalable One-Pass Optimisation of High-Dimensional Weight-Update Hyperparameters by Implicit Differentiation"

This repository contains code for the paper Scalable One-Pass Optimisation of High-Dimensional Weight-Update Hyperparameters by Implicit Differentiati

8 Aug 28, 2022
The Medical Detection Toolkit contains 2D + 3D implementations of prevalent object detectors such as Mask R-CNN, Retina Net, Retina U-Net, as well as a training and inference framework focused on dealing with medical images.

The Medical Detection Toolkit contains 2D + 3D implementations of prevalent object detectors such as Mask R-CNN, Retina Net, Retina U-Net, as well as a training and inference framework focused on dea

MIC-DKFZ 1.2k Jan 04, 2023