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
9 Dec 21, 2022
501 Dec 28, 2022
227 Dec 28, 2022
1.6k Dec 31, 2022
27 Sep 17, 2021
349 Dec 07, 2022
15 Feb 10, 2022
18 Aug 18, 2022
216 Dec 30, 2022
6.1k Jan 04, 2023
47 Dec 12, 2022
76 Dec 12, 2022
25 Jun 20, 2021
172 Dec 12, 2022
89 Jan 02, 2023
6 Oct 28, 2022
9 Aug 22, 2022
217 Dec 05, 2022
5 Jun 08, 2022
1 Dec 08, 2021