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

A pytorch-based real-time segmentation model for autonomous driving

Official implementation of NPMs: Neural Parametric Models for 3D Deformable Shapes - ICCV 2021

Semi-Supervised Signed Clustering Graph Neural Network (and Implementation of Some Spectral Methods)

A PyTorch implementation for Unsupervised Domain Adaptation by Backpropagation(DANN), support Office-31 and Office-Home dataset

Pytorch library for fast transformer implementations

On Nonlinear Latent Transformations for GAN-based Image Editing - PyTorch implementation

FridaHookAppTool - Frida Hook App Tool With Python

Exploring Relational Context for Multi-Task Dense Prediction [ICCV 2021]

[NeurIPS 2021 Spotlight] Aligning Pretraining for Detection via Object-Level Contrastive Learning

Python PID Tuner - Based on a FOPDT model obtained using a Open Loop Process Reaction Curve

Face recognition system using MTCNN, FACENET, SVM and FAST API to track participants of Big Brother Brasil in real time.

TraND: Transferable Neighborhood Discovery for Unsupervised Cross-domain Gait Recognition.

Implementation of paper "Decision-based Black-box Attack Against Vision Transformers via Patch-wise Adversarial Removal"

This repository includes the code of the sequence-to-sequence model for discontinuous constituent parsing described in paper Discontinuous Grammar as a Foreign Language.

Half Instance Normalization Network for Image Restoration

Diverse Branch Block: Building a Convolution as an Inception-like Unit

Fine-grained Control of Image Caption Generation with Abstract Scene Graphs

A PyTorch implementation of "Pathfinder Discovery Networks for Neural Message Passing"

This is the source code for: Context-aware Entity Typing in Knowledge Graphs.

VLG-Net: Video-Language Graph Matching Networks for Video Grounding