Transfer Learning of VGG19 trained using ImageNet an retrained for Cifar-10 Dataset using PyTorch.
Library | Version |
---|---|
Python | ^2.7 |
PyTorch | ^0.1.12 |
Numpy | ^1.12.0 |
Pickle | * |
git clone https://github.com/eugenelet/PyTorch-Transfer-Learning-of-VGG19-for-Cifar-10-Dataset
cd PyTorch-Transfer-Learning-of-VGG19-for-Cifar-10-Dataset
Batch size: 128
Prediction made on per epoch basis.
161 epochs takes about 3h on GTX 1080.
python train.py
python test.py
python test_vgg19.py
tensorboard --logdir=./tensorboard