PyTorch models trained on CIFAR-10 dataset

• I modified TorchVision official implementation of popular CNN models, and trained those on CIFAR-10 dataset.
• I changed number of class, filter size, stride, and padding in the the original code so that it works with CIFAR-10.
• I also share the weights of these models, so you can just load the weights and use them.
• The code is highly re-producible and readable by using PyTorch-Lightning.

Statistics of supported models

Details report

Weight and Biases' details report for this project WandB Report

How To Cite

How to use pretrained models

Automatically download and extract the weights from Box (933 MB)

python train.py --download_weights 1

Or use Google Drive backup link (you have to download and extract manually)

Load model and run

from cifar10_models.vgg import vgg11_bn, vgg13_bn, vgg16_bn, vgg19_bn

# Untrained model
my_model = vgg11_bn()

# Pretrained model
my_model = vgg11_bn(pretrained=True)
my_model.eval() # for evaluation

If you use your own images, all models expect data to be in range [0, 1] then normalized by

mean = [0.4914, 0.4822, 0.4465]
std = [0.2471, 0.2435, 0.2616]

How to train models from scratch

Check the train.py to see all available hyper-parameter choices. To reproduce the same accuracy use the default hyper-parameters

python train.py --classifier resnet18

How to test pretrained models

python train.py --test_phase 1 --pretrained 1 --classifier resnet18

Output

{'acc/test': tensor(93.0689, device='cuda:0')}

Requirements

Just to use pretrained models

• pytorch = 1.7.0

To train & test

• pytorch = 1.7.0
• torchvision = 0.7.0
• tensorboard = 2.2.1
• pytorch-lightning = 1.1.0