Downloads · 30 days
6
18% of all-time downloads
microvit-s3
henriquequeirozcunha/microvit-s3
microvit-s3 is a image classification model from henriquequeirozcunha. Use it when you need a label for an image. It is set up for transformers. The card lists the license as apache-2.0.
ImageNet-1K pretrained MicroViT-S3 (16.7M params, 77.1% Top-1 accuracy).
All-time downloads
33
Public
Parameters
14.3M
115 MB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pth57.4 MB · 50%
At a glance
- Task
- Image Classification
- Library
- transformers
- License
- apache-2.0
- Model type
- microvit
- Access
- Public
- Created
- Jul 1, 2026
- Updated
- Jul 1, 2026
- SHA
- dfe60ba4
Model card
From the Hugging Face model README
MicroViT-S3
ImageNet-1K pretrained MicroViT-S3 (16.7M params, 77.1% Top-1 accuracy).
Architecture: SHViT (Single-Head Vision Transformer) backbone. Source paper: https://arxiv.org/abs/2502.05800 Official repo: https://github.com/novendrastywn/MicroViT
Usage
from transformers import AutoModelForImageClassification, AutoImageProcessor
# Load ImageNet pretrained (1000 classes)
model = AutoModelForImageClassification.from_pretrained(
"henriquequeirozcunha/microvit-s3",
trust_remote_code=True,
)
# Fine-tune for binary classification
model = AutoModelForImageClassification.from_pretrained(
"henriquequeirozcunha/microvit-s3",
num_labels=2,
ignore_mismatched_sizes=True,
trust_remote_code=True,
)
Preprocessing: 224×224, ImageNet normalization (mean=[0.485,0.456,0.406], std=[0.229,0.224,0.225]).
- Task
- Image Classification
- Library
- transformers
- Type
- microvit
- License
- apache-2.0
- Created
- Jul 1, 2026
- Updated
- Jul 1, 2026