Lemonzhoumeng / pseudo_label-pytorch

The Simple and Efficient Semi-Supervised Learning Method for Deep Neural Networks

Pseudo-Label The Simple and Efficient Semi-Supervised Learning Method for Deep Neural Networks

The repository implement a semi-supervised method for Deep Neural Networks, the Pseudo Label. More details for the method please refer to Pseudo-Label The Simple and Efficient Semi-Supervised Learning Method for Deep Neural Networks.

The environment:

• Python 3.6.5 :: Anaconda
• PyTorch 0.4.0
• torchvision 0.2.1
• tensorboardX (for log)
• tensorflow (for visualization)

To prepare the data:

bash data-local/bin/prepare_cifar10.sh

To run the code:

python -m experiments.cifar10_test

Visualization:

Make sure you have installed the tensorflow for tensorboard

tensorboard --logdir runs

Code Reference

pytorch-cifar@kuangliu

mean-teacher@CuriousAI

senet.pytorch@moskomule