SSFormers

KO | EN

gitlite — search

#hacktoberfest #react #ai #typescript #golang #android #machine-learning #deep-learning #docker #nodejs #pytorch #security #openai #open-source

(★ 48)

This repository is the code of the paper "Sparse Spatial Transformers for Few-Shot Learning" (SCIENCE CHINA Information Sciences).

SSFormers Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# :star2: Sparse Spatial Transformers for Few-Shot Learning
This code implements the [Sparse Spatial Transformers for Few-Shot Learning(SSFormers)](https://arxiv.org/abs/2109.12932).

[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/sparse-spatial-transformers-for-few-shot/few-shot-image-classification-on-fc100-5-way)](https://paperswithcode.com/sota/few-shot-image-classification-on-fc100-5-way?p=sparse-spatial-transformers-for-few-shot)
[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/sparse-spatial-transformers-for-few-shot/few-shot-image-classification-on-fc100-5-way-1)](https://paperswithcode.com/sota/few-shot-image-classification-on-fc100-5-way-1?p=sparse-spatial-transformers-for-few-shot)
[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/sparse-spatial-transformers-for-few-shot/few-shot-image-classification-on-cifar-fs-5)](https://paperswithcode.com/sota/few-shot-image-classification-on-cifar-fs-5?p=sparse-spatial-transformers-for-few-shot)
[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/sparse-spatial-transformers-for-few-shot/few-shot-image-classification-on-cifar-fs-5-1)](https://paperswithcode.com/sota/few-shot-image-classification-on-cifar-fs-5-1?p=sparse-spatial-transformers-for-few-shot)
[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/sparse-spatial-transformers-for-few-shot/few-shot-image-classification-on-mini-3)](https://paperswithcode.com/sota/few-shot-image-classification-on-mini-3?p=sparse-spatial-transformers-for-few-shot)
[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/sparse-spatial-transformers-for-few-shot/few-shot-image-classification-on-mini-2)](https://paperswithcode.com/sota/few-shot-image-classification-on-mini-2?p=sparse-spatial-transformers-for-few-shot)
[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/sparse-spatial-transformers-for-few-shot/few-shot-image-classification-on-tiered)](https://paperswithcode.com/sota/few-shot-image-classification-on-tiered?p=sparse-spatial-transformers-for-few-shot)
[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/sparse-spatial-transformers-for-few-shot/few-shot-image-classification-on-tiered-1)](https://paperswithcode.com/sota/few-shot-image-classification-on-tiered-1?p=sparse-spatial-transformers-for-few-shot)

## :bookmark: Citation
If you find our work useful, please consider citing our work using the bibtex:
```
@Article{ssformers,
	author  = {Chen, Haoxing and Li, Huaxiong and Li, Yaohui and Chen, Chunlin},
	title   = {Sparse Spatial Transformers for Few-Shot Learning},
	journal = {Sci. China Inf. Sci.},
	year    = {2023},
}
```

## :palm_tree: Prerequisites
* Linux
* Python 3.8
* Pytorch 1.9.1
* GPU + CUDA CuDNN
* pillow, torchvision, scipy, numpy
## :bookmark_tabs: Datasets
**Dataset download link:**
* [miniImageNet](https://drive.google.com/file/d/1fUBrpv8iutYwdL4xE1rX_R9ef6tyncX9/view) It contains 100 classes with 600 images in each class, which are built upon the ImageNet dataset. The 100 classes are divided into 64, 16, 20 for meta-training, meta-validation and meta-testing, respectively.
* [tieredImageNet](https://drive.google.com/drive/folders/163HGKZTvfcxsY96uIF6ILK_6ZmlULf_j?usp=sharing)
TieredImageNet is also a subset of ImageNet, which includes 608 classes from 34 super-classes. Compared with miniImageNet, the splits of meta-training(20), meta-validation(6) and meta-testing(8) are set according to the super-classes to enlarge the domain difference between training and testing phase. The dataset also include more images for training and evaluation (779,165 images in total).
* [CIFAR-FS](https://drive.google.com/file/d/1GjGMI0q3bgcpcB_CjI40fX54WgLPuTpS/view?usp=sharing)
CIFAR-FS is divided from CIFAR-100, which consists of 60,000 images in 100 categories. The CIFAR-FS is divided into 64, 16 and 20 for training, validation, and evaluation, respectively.
* [FC-100](https://drive.google.com/file/d/1_ZsLyqI487NRDQhwvI7rg86FK3YAZvz1/view?usp=sharing)
FC-100 is also divided from CIFAR-100, which is more difficult because it is more diverse. The FC100 uses a split similar to tieredImageNet, where train, validation, and test splits contain 60, 20, and 20 classes.
**Note: You need to manually change the dataset directory.**

## Pre-trained backbone
We provide pre-trained backbones at https://pan.baidu.com/s/1v2k-mdCpGLtKnKG5ijYXMw  keys: 334q

## :four_leaf_clover: Few-shot Classification
* Train a 5-way 1-shot SSFormers model based on Conv-64F (on miniImageNet dataset):
```
 python experiments/run_trainer.py  --cfg ./configs/miniImagenet/ST_N5K1_R12.yaml --device 0
```
Test model on the test set:
```
python experiments/run_evaluator.py --cfg ./configs/miniImagenet/ST_N5K1_R12.yaml -c ./checkpoint/*/*.pth --device 0
```
and semi-supervised few-shot learning tasks (with trial t=1).
```
python experiments/run_semi_trainer.py --cfg ./configs/miniImagenet/ST_N5K1_semi_with_extractor.yaml --device 0 -t 1

python experiments/run_semi_evaluator.py --cfg ./configs/miniImagenet/ST_N5K1_semi_with_extractor.yaml -c ./checkpoints/*/*.pth --device 0
```
### Our code is based on [MCL](https://github.com/LouieYang/MCL) and [FEAT](https://github.com/Sha-Lab/FEAT).

## :email: Contacts
Please feel free to contact us if you have any problems.

Email: [haoxingchen@smail.nju.edu.cn](haoxingchen@smail.nju.edu.cn)

SSFormers

SSFormers Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

SSFormers

SSFormers Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README