|
--- |
|
license: apache-2.0 |
|
datasets: |
|
- uoft-cs/cifar100 |
|
language: |
|
- en |
|
metrics: |
|
- accuracy |
|
pipeline_tag: image-classification |
|
--- |
|
# Model Card for LGViT |
|
|
|
The checkpoint of LGViT using the DeiT backbone in the Cifar-100 dataset. |
|
|
|
## Model Details |
|
|
|
|
|
### Model Sources |
|
|
|
<!-- Provide the basic links for the model. --> |
|
|
|
- **Repository:** https://github.com/falcon-xu/LGViT |
|
- **Paper:** https://arxiv.org/abs/2308.00255 |
|
|
|
|
|
|
|
#### Metrics |
|
|
|
- Classification accuracy |
|
- Speed-Up |
|
|
|
### Results |
|
|
|
[More Information Needed] |
|
|
|
#### Summary |
|
|
|
|
|
|
|
## Citation |
|
|
|
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. --> |
|
|
|
**BibTeX:** |
|
|
|
``` |
|
@article{xu2023lgvit, |
|
title={LGViT: Dynamic Early Exiting for Accelerating Vision Transformer}, |
|
author={Xu, Guanyu and Hao, Jiawei and Shen, Li and Hu, Han and Luo, Yong and Lin, Hui and Shen, Jialie}, |
|
journal={arXiv preprint arXiv:2308.00255}, |
|
year={2023} |
|
} |
|
``` |
|
|
|
|
