https://github.com/fcakyon/cifar100-resnet
ResNet Implementation for CIFAR100 in Pytorch
https://github.com/fcakyon/cifar100-resnet
cifar cifar100 evaluate jupyter python pytorch resnet train
Last synced: 6 months ago
JSON representation
ResNet Implementation for CIFAR100 in Pytorch
- Host: GitHub
- URL: https://github.com/fcakyon/cifar100-resnet
- Owner: fcakyon
- License: mit
- Created: 2021-01-04T22:52:44.000Z (over 4 years ago)
- Default Branch: main
- Last Pushed: 2021-01-04T23:07:06.000Z (over 4 years ago)
- Last Synced: 2025-04-23T08:16:26.931Z (6 months ago)
- Topics: cifar, cifar100, evaluate, jupyter, python, pytorch, resnet, train
- Language: Jupyter Notebook
- Homepage:
- Size: 138 KB
- Stars: 5
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# ResNet Implementation for CIFAR100 in Pytorch
[Torchvision model zoo](https://github.com/pytorch/vision/tree/master/torchvision/models) provides number of implementations of various state-of-the-art architectures, however, most of them are defined and implemented for ImageNet. Usage of these backbones with small input sizes (as in CIFAR) is not trivial.
This repo provides training and evaluation scripts for CIFAR100 with ResNet backbones. 60% accuracy can be obtained with default training parameters.
Check [the notebook]() for a demo of the submodules used in this repo.
## Usage
- Clone:
```bash
git clone https://github.com/fcakyon/cifar100-resnet.git
```
- Prepare conda environment:
```bash
conda env create -f environment.yml
```
```bash
conda activate cifarresnet
```
- Train:
```bash
python main.py --arch resnet32 --save-dir checkpoints/
```
- Evaluate:
```bash
python main.py --evaluate checkpoints/resnet32_final.th
```
## Related Projects
[pytorch_resnet_cifar10](https://github.com/akamaster/pytorch_resnet_cifar10)