Official Implementation of SWAD (NeurIPS 2021)

Related tags

Deep Learningswad
Overview

SWAD: Domain Generalization by Seeking Flat Minima (NeurIPS'21)

Official PyTorch implementation of SWAD: Domain Generalization by Seeking Flat Minima.

Junbum Cha, Sanghyuk Chun, Kyungjae Lee, Han-Cheol Cho, Seunghyun Park, Yunsung Lee, Sungrae Park.

Note that this project is built upon [email protected].

Preparation

Dependencies

pip install -r requirements.txt

Datasets

python -m domainbed.scripts.download --data_dir=/my/datasets/path

Environments

Environment details used for our study.

Python: 3.8.6
PyTorch: 1.7.0+cu92
Torchvision: 0.8.1+cu92
CUDA: 9.2
CUDNN: 7603
NumPy: 1.19.4
PIL: 8.0.1

How to Run

train_all.py script conducts multiple leave-one-out cross-validations for all target domain.

python train_all.py exp_name --dataset PACS --data_dir /my/datasets/path

Experiment results are reported as a table. In the table, the row SWAD indicates out-of-domain accuracy from SWAD. The row SWAD (inD) indicates in-domain validation accuracy.

Example results:

+------------+--------------+---------+---------+---------+---------+
| Selection  | art_painting | cartoon |  photo  |  sketch |   Avg.  |
+------------+--------------+---------+---------+---------+---------+
|   oracle   |   82.245%    | 85.661% | 97.530% | 83.461% | 87.224% |
|    iid     |   87.919%    | 78.891% | 96.482% | 78.435% | 85.432% |
|    last    |   82.306%    | 81.823% | 95.135% | 82.061% | 85.331% |
| last (inD) |   95.807%    | 95.291% | 96.306% | 95.477% | 95.720% |
| iid (inD)  |   97.275%    | 96.619% | 96.696% | 97.253% | 96.961% |
|    SWAD    |   89.750%    | 82.942% | 97.979% | 81.870% | 88.135% |
| SWAD (inD) |   97.713%    | 97.649% | 97.316% | 98.074% | 97.688% |
+------------+--------------+---------+---------+---------+---------+

In this example, the DG performance of SWAD for PACS dataset is 88.135%.

If you set indomain_test option to True, the validation set is splitted to validation and test sets, and the (inD) keys become to indicate in-domain test accuracy.

Reproduce the results of the paper

We provide the instructions to reproduce the main results of the paper, Table 1 and 2. Note that the difference in a detailed environment or uncontrolled randomness may bring a little different result from the paper.

  • PACS
python train_all.py PACS0 --dataset PACS --deterministic --trial_seed 0 --checkpoint_freq 100 --data_dir /my/datasets/path
python train_all.py PACS1 --dataset PACS --deterministic --trial_seed 1 --checkpoint_freq 100 --data_dir /my/datasets/path
python train_all.py PACS2 --dataset PACS --deterministic --trial_seed 2 --checkpoint_freq 100 --data_dir /my/datasets/path
  • VLCS
python train_all.py VLCS0 --dataset VLCS --deterministic --trial_seed 0 --checkpoint_freq 50 --tolerance_ratio 0.2 --data_dir /my/datasets/path
python train_all.py VLCS1 --dataset VLCS --deterministic --trial_seed 1 --checkpoint_freq 50 --tolerance_ratio 0.2 --data_dir /my/datasets/path
python train_all.py VLCS2 --dataset VLCS --deterministic --trial_seed 2 --checkpoint_freq 50 --tolerance_ratio 0.2 --data_dir /my/datasets/path
  • OfficeHome
python train_all.py OH0 --dataset OfficeHome --deterministic --trial_seed 0 --checkpoint_freq 100 --data_dir /my/datasets/path
python train_all.py OH1 --dataset OfficeHome --deterministic --trial_seed 1 --checkpoint_freq 100 --data_dir /my/datasets/path
python train_all.py OH2 --dataset OfficeHome --deterministic --trial_seed 2 --checkpoint_freq 100 --data_dir /my/datasets/path
  • TerraIncognita
python train_all.py TR0 --dataset TerraIncognita --deterministic --trial_seed 0 --checkpoint_freq 100 --data_dir /my/datasets/path
python train_all.py TR1 --dataset TerraIncognita --deterministic --trial_seed 1 --checkpoint_freq 100 --data_dir /my/datasets/path
python train_all.py TR2 --dataset TerraIncognita --deterministic --trial_seed 2 --checkpoint_freq 100 --data_dir /my/datasets/path
  • DomainNet
python train_all.py DN0 --dataset DomainNet --deterministic --trial_seed 0 --checkpoint_freq 500 --data_dir /my/datasets/path
python train_all.py DN1 --dataset DomainNet --deterministic --trial_seed 1 --checkpoint_freq 500 --data_dir /my/datasets/path
python train_all.py DN2 --dataset DomainNet --deterministic --trial_seed 2 --checkpoint_freq 500 --data_dir /my/datasets/path

Main Results

Citation

The paper will be published at NeurIPS 2021.

@inproceedings{cha2021swad,
  title={SWAD: Domain Generalization by Seeking Flat Minima},
  author={Cha, Junbum and Chun, Sanghyuk and Lee, Kyungjae and Cho, Han-Cheol and Park, Seunghyun and Lee, Yunsung and Park, Sungrae},
  booktitle={Advances in Neural Information Processing Systems (NeurIPS)},
  year={2021}
}

License

This source code is released under the MIT license, included here.

This project includes some code from DomainBed, also MIT licensed.

Source code to accompany Defunctland's video "FASTPASS: A Complicated Legacy"

Shapeland Simulator Source code to accompany Defunctland's video "FASTPASS: A Complicated Legacy" Download the video at https://www.youtube.com/watch?

TouringPlans.com 70 Dec 14, 2022
A Python Package for Portfolio Optimization using the Critical Line Algorithm

PyCLA A Python Package for Portfolio Optimization using the Critical Line Algorithm Getting started To use PyCLA, clone the repo and install the requi

19 Oct 11, 2022
Codes accompanying the paper "Believe What You See: Implicit Constraint Approach for Offline Multi-Agent Reinforcement Learning" (NeurIPS 2021 Spotlight

Implicit Constraint Q-Learning This is a pytorch implementation of ICQ on Datasets for Deep Data-Driven Reinforcement Learning (D4RL) and ICQ-MA on SM

42 Dec 23, 2022
Picasso: a methods for embedding points in 2D in a way that respects distances while fitting a user-specified shape.

Picasso Code to generate Picasso embeddings of any input matrix. Picasso maps the points of an input matrix to user-defined, n-dimensional shape coord

Pachter Lab 45 Dec 23, 2022
Boosted neural network for tabular data

XBNet - Xtremely Boosted Network Boosted neural network for tabular data XBNet is an open source project which is built with PyTorch which tries to co

Tushar Sarkar 175 Jan 04, 2023
Sum-Product Probabilistic Language

Sum-Product Probabilistic Language SPPL is a probabilistic programming language that delivers exact solutions to a broad range of probabilistic infere

MIT Probabilistic Computing Project 57 Nov 17, 2022
My personal Home Assistant configuration.

About This is my personal Home Assistant configuration. My guiding princile is to have full local control of all my devices. I intend everything to ru

Chris Turra 13 Jun 07, 2022
IndoNLI: A Natural Language Inference Dataset for Indonesian

IndoNLI: A Natural Language Inference Dataset for Indonesian This is a repository for data and code accompanying our EMNLP 2021 paper "IndoNLI: A Natu

15 Feb 10, 2022
METS/ALTO OCR enhancing tool by the National Library of Luxembourg (BnL)

Nautilus-OCR The National Library of Luxembourg (BnL) started its first initiative in digitizing newspapers, with layout recognition and OCR on articl

National Library of Luxembourg 36 Dec 05, 2022
Official code for "InfoGraph: Unsupervised and Semi-supervised Graph-Level Representation Learning via Mutual Information Maximization" (ICLR 2020, spotlight)

InfoGraph: Unsupervised and Semi-supervised Graph-Level Representation Learning via Mutual Information Maximization Authors: Fan-yun Sun, Jordan Hoffm

Fan-Yun Sun 232 Dec 28, 2022
Generic U-Net Tensorflow implementation for image segmentation

Tensorflow Unet Warning This project is discontinued in favour of a Tensorflow 2 compatible reimplementation of this project found under https://githu

Joel Akeret 1.8k Dec 10, 2022
Source code for Zalo AI 2021 submission

zalo_ltr_2021 Source code for Zalo AI 2021 submission Solution: Pipeline We use the pipepline in the picture below: Our pipeline is combination of BM2

128 Dec 27, 2022
Official implementation of VaxNeRF (Voxel-Accelearated NeRF).

VaxNeRF Paper | Google Colab This is the official implementation of VaxNeRF (Voxel-Accelearated NeRF). VaxNeRF provides very fast training and slightl

naruya 132 Nov 21, 2022
AutoML library for deep learning

Official Website: autokeras.com AutoKeras: An AutoML system based on Keras. It is developed by DATA Lab at Texas A&M University. The goal of AutoKeras

Keras 8.7k Jan 08, 2023
Official implementation of "Membership Inference Attacks Against Self-supervised Speech Models"

Introduction Official implementation of "Membership Inference Attacks Against Self-supervised Speech Models". In this work, we demonstrate that existi

Wei-Cheng Tseng 7 Nov 01, 2022
Official implementation for Multi-Modal Interaction Graph Convolutional Network for Temporal Language Localization in Videos

Multi-modal Interaction Graph Convolutioal Network for Temporal Language Localization in Videos Official implementation for Multi-Modal Interaction Gr

Zongmeng Zhang 15 Oct 18, 2022
Code for ACL 2019 Paper: "COMET: Commonsense Transformers for Automatic Knowledge Graph Construction"

To run a generation experiment (either conceptnet or atomic), follow these instructions: First Steps First clone, the repo: git clone https://github.c

Antoine Bosselut 575 Jan 01, 2023
A new play-and-plug method of controlling an existing generative model with conditioning attributes and their compositions.

Viz-It Data Visualizer Web-Application If I ask you where most of the data wrangler looses their time ? It is Data Overview and EDA. Presenting "Viz-I

NVIDIA Research Projects 66 Jan 01, 2023
An Empirical Investigation of Model-to-Model Distribution Shifts in Trained Convolutional Filters

CNN-Filter-DB An Empirical Investigation of Model-to-Model Distribution Shifts in Trained Convolutional Filters Paul Gavrikov, Janis Keuper Paper: htt

Paul Gavrikov 18 Dec 30, 2022
implicit displacement field

Geometry-Consistent Neural Shape Representation with Implicit Displacement Fields [project page][paper][cite] Geometry-Consistent Neural Shape Represe

Yifan Wang 100 Dec 19, 2022