Original : https://github.com/kuangliu/pytorch-cifar by kuangliu.
- Python 3.6+
- PyTorch 1.0+
Add logger and checkpoint. Log and checkpoint files will be saved into each directory. (./log/[cur_time] && ./checkpoint/[cur_time])
You can adjust learning rate with manual method or auto method.
Train with auto scheduler with python3 main.py --scheduler
or you can use a manual way with adjust_learning_rate().
With python3 main.py --help, you will get more information.
| Model | Acc. |
|---|---|
| VGG16 | 92.64% |
| ResNet18 | 93.02% |
| ResNet50 | 93.62% |
| ResNet101 | 93.75% |
| MobileNetV2 | 94.43% |
| ResNeXt29(32x4d) | 94.73% |
| ResNeXt29(2x64d) | 94.82% |
| DenseNet121 | 95.04% |
| PreActResNet18 | 95.11% |
| DPN92 | 95.16% |