Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/apple/ml-mobileone
This repository contains the official implementation of the research paper, "An Improved One millisecond Mobile Backbone".
https://github.com/apple/ml-mobileone
Last synced: 27 days ago
JSON representation
This repository contains the official implementation of the research paper, "An Improved One millisecond Mobile Backbone".
- Host: GitHub
- URL: https://github.com/apple/ml-mobileone
- Owner: apple
- License: other
- Created: 2022-07-13T18:29:33.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2022-07-25T21:19:45.000Z (over 2 years ago)
- Last Synced: 2024-09-29T21:41:22.560Z (about 1 month ago)
- Language: Swift
- Homepage:
- Size: 731 KB
- Stars: 723
- Watchers: 17
- Forks: 61
- Open Issues: 19
-
Metadata Files:
- Readme: README.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
- Code of conduct: CODE_OF_CONDUCT.md
Awesome Lists containing this project
README
# MobileOne: An Improved One millisecond Mobile Backbone
This software project accompanies the research paper, [An Improved One millisecond Mobile Backbone](https://arxiv.org/abs/2206.04040).
Our model achieves Top-1 Accuracy of 75.9\% under 1ms.
## Model Zoo
### ImageNet-1K
| Model | Top-1 Acc.| Latency* | Pytorch Checkpoint (url) | CoreML Model |
| :--- | :---: | :---: | :---: | :---: |
| MobileOne-S0 | 71.4 | 0.79 | [S0](https://docs-assets.developer.apple.com/ml-research/datasets/mobileone/mobileone_s0.pth.tar)([unfused](https://docs-assets.developer.apple.com/ml-research/datasets/mobileone/mobileone_s0_unfused.pth.tar)) | [mlmodel](https://docs-assets.developer.apple.com/ml-research/datasets/mobileone/mobileone_s0.mlmodel) |
| MobileOne-S1 | 75.9 | 0.89 | [S1](https://docs-assets.developer.apple.com/ml-research/datasets/mobileone/mobileone_s1.pth.tar)([unfused](https://docs-assets.developer.apple.com/ml-research/datasets/mobileone/mobileone_s1_unfused.pth.tar)) | [mlmodel](https://docs-assets.developer.apple.com/ml-research/datasets/mobileone/mobileone_s1.mlmodel) |
| MobileOne-S2 | 77.4 | 1.18 | [S2](https://docs-assets.developer.apple.com/ml-research/datasets/mobileone/mobileone_s2.pth.tar)([unfused](https://docs-assets.developer.apple.com/ml-research/datasets/mobileone/mobileone_s2_unfused.pth.tar)) | [mlmodel](https://docs-assets.developer.apple.com/ml-research/datasets/mobileone/mobileone_s2.mlmodel) |
| MobileOne-S3 | 78.1 | 1.53 | [S3](https://docs-assets.developer.apple.com/ml-research/datasets/mobileone/mobileone_s3.pth.tar)([unfused](https://docs-assets.developer.apple.com/ml-research/datasets/mobileone/mobileone_s3_unfused.pth.tar)) | [mlmodel](https://docs-assets.developer.apple.com/ml-research/datasets/mobileone/mobileone_s3.mlmodel) |
| MobileOne-S4 | 79.4 | 1.86 | [S4](https://docs-assets.developer.apple.com/ml-research/datasets/mobileone/mobileone_s4.pth.tar)([unfused](https://docs-assets.developer.apple.com/ml-research/datasets/mobileone/mobileone_s4_unfused.pth.tar)) | [mlmodel](https://docs-assets.developer.apple.com/ml-research/datasets/mobileone/mobileone_s4.mlmodel) |
*Latency measured on iPhone 12 Pro.
## Usage
To use our model, follow the code snippet below,
```python
import torch
from mobileone import mobileone, reparameterize_model
# To Train from scratch/fine-tuning
model = mobileone(variant='s0')
# ... train ...
# Load Pre-trained checkpoint for fine-tuning
checkpoint = torch.load('/path/to/unfused_checkpoint.pth.tar')
model.load_state_dict(checkpoint)
# ... train ...
# For inference
model.eval()
model_eval = reparameterize_model(model)
# Use model_eval at test-time
```
To simply evaluate our model, use the fused checkpoint where branches are re-parameterized.
```python
import torch
from mobileone import mobileone
model = mobileone(variant='s0', inference_mode=True)
checkpoint = torch.load('/path/to/checkpoint.pth.tar')
model.load_state_dict(checkpoint)
# ... evaluate/demo ...
```
## ModelBench App
An iOS benchmark app for MobileOne CoreML models. See
[ModelBench](ModelBench/README.md) for addition details on building
and running the app.
## Citation
If our code or models help your work, please cite our [paper](https://arxiv.org/abs/2206.04040):
```BibTeX
@article{mobileone2022,
title={An Improved One millisecond Mobile Backbone},
author={Vasu, Pavan Kumar Anasosalu and Gabriel, James and Zhu, Jeff and Tuzel, Oncel and Ranjan, Anurag},
journal={arXiv preprint arXiv:2206.04040},
year={2022}
}
```