/CAMtrast

code for AAAI2022 paper 'Boost Supervised Pretraining for Visual Transfer Learning: Implications of Self-Supervised Contrastive Representation Learning'.

Primary LanguagePython

Boost Supervised Pretraining for Visual Transfer Learning: Implications of Self-Supervised Contrastive Representation Learning

This repo contains the reference source code for the paper [Boost Supervised Pretraining for Visual Transfer Learning: Implications of Self-Supervised Contrastive Representation Learning] in AAAI2021. We also provided supplementary materials. Our implementation is based on Pytorch.

Editor

This repository was built off of Contrastive Multiview Coding.

Run the code

Train

python3.6 -u train.py --epochs 100 --batch_size 256 --num_workers 24 --nce_k 2048 --softmax --model resnet50st --aug cjv2 --model_name [model_name] --n_way 64 --epoch_t 30

Path flags:

  • --model_name: specify the name to save model.

Test

python3 test.py --resume [resume] [data_folder] --gpu 1 --arch resnet50st --n_way 5 --k_shot 5 --task_num 600 --moco-k 2048 -j 8 --train_way 64

Path flags:

  • --resume: specify the path of pretrained model.
  • data_folder: specify the data folder.

Citation

Please cite our paper if the code is helpful to your research.

Concact

If you have any question, please feel free to concat Jinghan Sun (Email: jhsun@stu.xmu.edu.cn)