https://github.com/Randl/ShuffleNetV2-pytorch

Implementation of ShuffleNetV2 for pytorch
https://github.com/Randl/ShuffleNetV2-pytorch

cnn-model pytorch shufflenetv2

Last synced: about 2 months ago
JSON representation

Implementation of ShuffleNetV2 for pytorch

• Host: GitHub
• URL: https://github.com/Randl/ShuffleNetV2-pytorch
• Owner: Randl
• License: mit
• Created: 2018-08-02T13:06:48.000Z (almost 6 years ago)
• Default Branch: master
• Last Pushed: 2022-04-17T04:44:41.000Z (about 2 years ago)
• Last Synced: 2024-02-04T22:40:15.998Z (5 months ago)
• Topics: cnn-model, pytorch, shufflenetv2
• Language: Python
• Size: 10.2 MB
• Stars: 185
• Watchers: 9
• Forks: 52
• Open Issues: 5
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Lists

• awesome-image-classification - unofficial-pytorch : https://github.com/Randl/ShuffleNetV2-pytorch
• awesome-image-classification - unofficial-pytorch : https://github.com/Randl/ShuffleNetV2-pytorch

README

        
# ShuffleNetv2 in PyTorch

An implementation of `ShuffleNetv2` in PyTorch. `ShuffleNetv2` is an efficient convolutional neural network architecture for mobile devices. For more information check the paper:

[ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design](https://arxiv.org/abs/1807.11164)

## Usage

Clone the repo:

```bash

git clone https://github.com/Randl/ShuffleNetV2-pytorch

pip install -r requirements.txt

```

Use the model defined in `model.py` to run ImageNet example:

```bash

python imagenet.py --dataroot "/path/to/imagenet/"

```

To continue training from checkpoint

```bash

python imagenet.py --dataroot "/path/to/imagenet/" --resume "/path/to/checkpoint/folder"

```

## Results

For x0.5 model I achieved 0.4% lower top-1 accuracy than claimed.

|Classification Checkpoint| MACs (M)   | Parameters (M)| Top-1 Accuracy| Top-5 Accuracy|  Claimed top-1|  Claimed top-5|

|-------------------------|------------|---------------|---------------|---------------|---------------|---------------|

|      [shufflenet_v2_0.5]|41          |1.37           |          59.86|          81.63|           60.3|              -|

You can test it with

```bash

python imagenet.py --dataroot "/path/to/imagenet/" --resume "results/shufflenet_v2_0.5/model_best.pth.tar" -e --scaling 0.5

```