Train CIFAR10 with PyTorch
I'm playing with PyTorch on the CIFAR10 dataset.
Revision Logs:
- May 27: Append the zero grad measurement for this repo. Mainly to show the non-zero gradients rates
Prerequisites
- Python 3.6+
- PyTorch 1.0+
Training
# Start training with:
python main.py
# You can manually resume the training with:
python main.py --resume --lr=0.01
Accuracy
| Model | Acc. | Zero_Rate |
|---|---|---|
| VGG16 | 92.64% | |
| ResNet18 | 93.02% | |
| ResNet50 | 93.62% | |
| ResNet101 | 93.75% | |
| RegNetX_200MF | 94.24% | |
| RegNetY_400MF | 94.29% | |
| MobileNetV2 | 94.43% | |
| ResNeXt29(32x4d) | 94.73% | |
| ResNeXt29(2x64d) | 94.82% | |
| SimpleDLA | 94.89% | |
| DenseNet121 | 95.04% | |
| PreActResNet18 | 95.11% | |
| DPN92 | 95.16% | |
| DLA | 95.47% | |
| [AlexNet_with_lr=0.1] | 81.97% | 85.22% |