https://github.com/eveningdong/DAOSL

Implementation of Domain Adaption in One-Shot Learning
https://github.com/eveningdong/DAOSL

domain-adaptation one-shot-learning reinforcement-learning slim tensorflow

Last synced: about 2 months ago
JSON representation

Implementation of Domain Adaption in One-Shot Learning

• Host: GitHub
• URL: https://github.com/eveningdong/DAOSL
• Owner: eveningdong
• Created: 2018-11-12T03:40:56.000Z (about 6 years ago)
• Default Branch: master
• Last Pushed: 2019-03-11T03:56:21.000Z (almost 6 years ago)
• Last Synced: 2024-08-10T07:01:17.127Z (4 months ago)
• Topics: domain-adaptation, one-shot-learning, reinforcement-learning, slim, tensorflow
• Language: Python
• Homepage:
• Size: 13.9 MB
• Stars: 15
• Watchers: 2
• Forks: 6
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

• awesome-few-shot-meta-learning - code - official (TF)

README

        
# Domain Adaption in One-Shot Learning

Implementation of Domain Adaption in One-Shot Learning.

## Paper

You can find our paper at [Springer](https://link.springer.com/chapter/10.1007/978-3-030-10925-7_35) or [PDF](https://github.com/NanqingD/DAOSL/blob/master/ECML_2018_Camera_Ready_Final.pdf).

## Citation

If you find DAOSL useful in your research, please consider to cite:

@inproceedings{dong2018domain, 

  title={Domain adaption in one-shot learning}, 

  author={Dong, Nanqing and Eric P. Xing}, 

  booktitle={Joint European Conference on Machine Learning and Knowledge Discovery in Databases},

  pages={573--588},

  year={2018},

  organization={Springer}

}

## Setup

### Requirements

```

python 3.5

tensorflow 1.8

CUDA  9.0

cuDNN 7.0

```

### Installation

```

sh setup.sh

```

## Training

Train one-shot classifier for 5-way 1-shot learning.

```

python3 convert_data.py --data-name=omniglot

python3 convert_data.py --data-name=emnist --num-target-examples=20

python3 train_one_shot.py --exp-name=one_shot --source=omniglot --target=emnist_20 --num-ways=5

```

Train adversarial domain adaption (ADA) for 5-way 1-shot learning.

```

python3 convert_data.py --data-name=omniglot

python3 convert_data.py --data-name=emnist --num-target-examples=20

python3 train_ada.py --exp-name=ada --source=omniglot --target=emnist_20 --num-ways=5 --la=0.001

```

Train adversarial domain adaption (ADA) with reinforced sample selection (RSS) for 5-way 1-shot learning.

```

python3 convert_data.py --data-name=chars

python3 convert_data.py --data-name=sim

python3 convert_data.py --data-name=dis

python3 train_rss.py --exp-name=rss --num-ways=5 --la=0.001 --gamma=0.1

```