CADA-VAE-PyTorch - GitLite

KO | EN

gitlite — search

#javascript #hacktoberfest #react #go #android #machine-learning #deep-learning #linux #macos #nodejs #java #cli #kubernetes #vue

CADA-VAE-PyTorch

(★ 295)

Official implementation of the paper "Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders" (CVPR 2019)

파일 탐색기

최종 버전 다운로드 (.zip)

- data_loader.py
- final_classifier.py
- models.py
- single_experiment.py
- vaemodel.py
.gitignore
cadavae.yaml
LICENSE
Model.jpg
README.md
requirements.txt

# 설치 가이드

1. 코드 내려받기

git clone https://github.com/edgarschnfld/CADA-VAE-PyTorch

깃허브에서 프로젝트 코드 전체를 내 컴퓨터로 내려받습니다.

cd CADA-VAE-PyTorch

방금 내려받은 프로젝트 폴더 안으로 이동합니다.

2. Python

쉬움 추천

사전 준비물

Git GitHub에서 프로젝트 코드를 내려받으려면 필요합니다.
Python 3 설치 시 'Add Python to PATH' 옵션을 꼭 체크하세요 (Windows).

pip install -r requirements.txt

requirements.txt 등에 명시된 파이썬 라이브러리를 설치합니다.

python <실행할 파일명>.py # README에서 정확한 실행 파일명을 확인하세요

파이썬 스크립트(또는 모듈)를 실행합니다.

✅ 에러 메시지 없이 실행되고 터미널에 안내 문구가 출력되면 정상입니다.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# CADA-VAE
Original PyTorch implementation of "Generalized Zero-and Few-Shot Learning via Aligned Variational Autoencoders" (CVPR 2019).

Paper: https://arxiv.org/pdf/1812.01784.pdf

<p align="center">
  <img width="600" src="Model.jpg">
</p>
<p align="justify">
  
### Requirements
The code was implemented using Python 3.5.6 and the following packages:
```
torch==0.4.1
numpy==1.14.3
scipy==1.1.0
scikit_learn==0.20.3
networkx==1.11
```
Using Python 2 is not recommended.

### Data
Download the following folder https://www.dropbox.com/sh/btoc495ytfbnbat/AAAaurkoKnnk0uV-swgF-gdSa?dl=0
and put it in this repository.
Next to the folder "model", there should be a folder "data".

### Experiments

To run the experiments from the paper, navigate to the model folder and execute the following:
```
python single_experiment.py --dataset CUB --num_shots 0 --generalized True
```
The choices for the input arguments are:
```
datasets: CUB, SUN, AWA1, AWA2
num_shots: any number 
generalized: True, False
```
More hyperparameters can be adjusted in the file single_experiment.py directly. The results vary by 1-2% between identical runs.

### Citation
If you use this work please cite
```
@inproceedings{schonfeld2019generalized,
  title={Generalized zero-and few-shot learning via aligned variational autoencoders},
  author={Schonfeld, Edgar and Ebrahimi, Sayna and Sinha, Samarth and Darrell, Trevor and Akata, Zeynep},
  booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
  pages={8247--8255},
  year={2019}
}
```
### Contact 
For questions or help, feel welcome to write an email to edgarschoenfeld@live.de