Improving Transferability of Representations via Augmentation-Aware Self-Supervision

Last update: Sep 16, 2022

Related tags

Overview

Improving Transferability of Representations via Augmentation-Aware Self-Supervision

Accepted to NeurIPS 2021

TL;DR: Learning augmentation-aware information by predicting the difference between two augmented samples improves the transferability of representations.

Dependencies

conda create -n AugSelf python=3.8 pytorch=1.7.1 torchvision=0.8.2 cudatoolkit=10.1 ignite -c pytorch
conda activate AugSelf
pip install scipy tensorboard kornia==0.4.1 sklearn

Checkpoints

We provide ImageNet100-pretrained models in this Dropbox link.

Pretraining

We here provide SimSiam+AugSelf pretraining scripts. For training the baseline (i.e., no AugSelf), remove --ss-crop and --ss-color options. For using other frameworks like SimCLR, use the --framework option.

STL-10

CUDA_VISIBLE_DEVICES=0 python pretrain.py \
    --logdir ./logs/stl10/simsiam/aug_self \
    --framework simsiam \
    --dataset stl10 \
    --datadir DATADIR \
    --model resnet18 \
    --batch-size 256 \
    --max-epochs 200 \
    --ss-color 1.0 --ss-crop 1.0

ImageNet100

python pretrain.py \
    --logdir ./logs/imagenet100/simsiam/aug_self \
    --framework simsiam \
    --dataset imagenet100 \
    --datadir DATADIR \
    --batch-size 256 \
    --max-epochs 500 \
    --model resnet50 \
    --base-lr 0.05 --wd 1e-4 \
    --ckpt-freq 50 --eval-freq 50 \
    --ss-crop 0.5 --ss-color 0.5 \
    --num-workers 16 --distributed

Evaluation

Our main evaluation setups are linear evaluation on fine-grained classification datasets (Table 1) and few-shot benchmarks (Table 2).

linear evaluation

CUDA_VISIBLE_DEVICES=0 python transfer_linear_eval.py \
    --pretrain-data imagenet100 \
    --ckpt CKPT \
    --model resnet50 \
    --dataset cifar10 \
    --datadir DATADIR \
    --metric top1

few-shot

CUDA_VISIBLE_DEVICES=0 python transfer_few_shot.py \
    --pretrain-data imagenet100 \
    --ckpt CKPT \
    --model resnet50 \
    --dataset cub200 \
    --datadir DATADIR

Improving Transferability of Representations via Augmentation-Aware Self-Supervision

Related tags

Overview

Improving Transferability of Representations via Augmentation-Aware Self-Supervision

Dependencies

Checkpoints

Pretraining

STL-10

ImageNet100

Evaluation

linear evaluation

few-shot

Owner

hankook

Implementation of the final project of the course DDA6309 Probabilistic Graphical Model

SMPLpix: Neural Avatars from 3D Human Models

A non-linear, non-parametric Machine Learning method capable of modeling complex datasets

利用Tensorflow实现基于CNN的中文短文本分类

[NeurIPS 2021 Spotlight] Code for Learning to Compose Visual Relations

Official pytorch implementation of "DSPoint: Dual-scale Point Cloud Recognition with High-frequency Fusion"

Torchlight2 lan game server tool - A message forwarding tool for Torchlight 2 lan game

Learned Initializations for Optimizing Coordinate-Based Neural Representations

ProFuzzBench - A Benchmark for Stateful Protocol Fuzzing

Exe-to-xlsm - Simple script to create VBscript of exe and inject to xlsm

Python implementation of Project Fluent

Generative Models for Graph-Based Protein Design

UniLM AI - Large-scale Self-supervised Pre-training across Tasks, Languages, and Modalities

A PyTorch implementation of the paper "Semantic Image Synthesis via Adversarial Learning" in ICCV 2017

LexGLUE: A Benchmark Dataset for Legal Language Understanding in English

Pytorch Implementation of rpautrat/SuperPoint

CUP-DNN is a deep neural network model used to predict tissues of origin for cancers of unknown of primary.

City-seeds - A random generator of cultural characteristics intended to spark ideas and help draw threads

PuppetGAN - Cross-Domain Feature Disentanglement and Manipulation just got way better! 🚀

Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising