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

PiRapGenerator - Make anyone rap the digits of pi

Kaggle DSTL Satellite Imagery Feature Detection

Weakly supervised medical named entity classification

A robust camera and Lidar fusion based velocity estimator to undistort the pointcloud.

Conversion between units used in magnetism

An implementation of a discriminant function over a normal distribution to help classify datasets.

Source codes for "Structure-Aware Abstractive Conversation Summarization via Discourse and Action Graphs"

This is the repo for our work "Towards Persona-Based Empathetic Conversational Models" (EMNLP 2020)

DANA paper supplementary materials

PyTorch implementation of D2C: Diffuison-Decoding Models for Few-shot Conditional Generation.

Simple streamlit app to demonstrate HERE Tour Planning

A PyTorch implementation of Learning to learn by gradient descent by gradient descent

Data from "HateCheck: Functional Tests for Hate Speech Detection Models" (Röttger et al., ACL 2021)

Optimized primitives for collective multi-GPU communication

Unsupervised Image Generation with Infinite Generative Adversarial Networks

Fast methods to work with hydro- and topography data in pure Python.

Dynamic Neural Representational Decoders for High-Resolution Semantic Segmentation

Human POSEitioning System (HPS): 3D Human Pose Estimation and Self-localization in Large Scenes from Body-Mounted Sensors, CVPR 2021

[ICCV 2021] Official Pytorch implementation for Discriminative Region-based Multi-Label Zero-Shot Learning SOTA results on NUS-WIDE and OpenImages

《Image2Reverb: Cross-Modal Reverb Impulse Response Synthesis》(2021)