https://github.com/YU1ut/MixMatch-pytorch

Code for "MixMatch - A Holistic Approach to Semi-Supervised Learning"
https://github.com/YU1ut/MixMatch-pytorch

deep-learning pytorch semi-supervised-learning

Last synced: about 2 months ago
JSON representation

Code for "MixMatch - A Holistic Approach to Semi-Supervised Learning"

• Host: GitHub
• URL: https://github.com/YU1ut/MixMatch-pytorch
• Owner: YU1ut
• License: mit
• Created: 2019-05-22T03:22:19.000Z (over 5 years ago)
• Default Branch: master
• Last Pushed: 2023-11-02T15:46:45.000Z (about 1 year ago)
• Last Synced: 2024-08-04T03:11:07.051Z (5 months ago)
• Topics: deep-learning, pytorch, semi-supervised-learning
• Language: Python
• Size: 20.5 KB
• Stars: 630
• Watchers: 12
• Forks: 129
• Open Issues: 27
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# MixMatch

This is an unofficial PyTorch implementation of [MixMatch: A Holistic Approach to Semi-Supervised Learning](https://arxiv.org/abs/1905.02249). 

The official Tensorflow implementation is [here](https://github.com/google-research/mixmatch).

Now only experiments on CIFAR-10 are available.

This repository carefully implemented important details of the official implementation to reproduce the results.

## Requirements

- Python 3.6+

- PyTorch 1.0

- **torchvision 0.2.2 (older versions are not compatible with this code)** 

- tensorboardX

- progress

- matplotlib

- numpy

## Usage

### Train

Train the model by 250 labeled data of CIFAR-10 dataset:

```

python train.py --gpu  --n-labeled 250 --out cifar10@250

```

Train the model by 4000 labeled data of CIFAR-10 dataset:

```

python train.py --gpu  --n-labeled 4000 --out cifar10@4000

```

### Monitoring training progress

```

tensorboard.sh --port 6006 --logdir cifar10@250

```

## Results (Accuracy)

| #Labels | 250 | 500 | 1000 | 2000| 4000 |

|:---|:---:|:---:|:---:|:---:|:---:|

|Paper | 88.92 ± 0.87 | 90.35 ± 0.94 | 92.25 ± 0.32| 92.97 ± 0.15 |93.76 ± 0.06|

|This code | 88.71 | 88.96 | 90.52 | 92.23 | 93.52 |

(Results of this code were evaluated on 1 run. Results of 5 runs with different seeds will be updated later. )

## References

```

@article{berthelot2019mixmatch,

  title={MixMatch: A Holistic Approach to Semi-Supervised Learning},

  author={Berthelot, David and Carlini, Nicholas and Goodfellow, Ian and Papernot, Nicolas and Oliver, Avital and Raffel, Colin},

  journal={arXiv preprint arXiv:1905.02249},

  year={2019}

}

```