Samvit Large Patch16.sa1b

🚀 samvit_large_patch16.sa1b

A Segment-Anything Vision Transformer (SAM ViT) image feature model (NOTE: for features and fine-tune, segmentation head not included). Pretrained on SA-1B for segmentation by paper authors w/ initialization from MAE weights.

🚀 Quick Start

This is a Segment-Anything Vision Transformer (SAM ViT) image feature model. It's designed for feature extraction and fine - tuning, without a segmentation head. It was pretrained on the SA - 1B dataset by the paper authors, initialized with MAE weights.

✨ Features

• Model Type: Image classification / feature backbone
• Model Stats:
  • Params (M): 308.3
  • GMACs: 1493.9
  • Activations (M): 2553.8
  • Image size: 1024 x 1024
• Papers:
  • Segment Anything
  • An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
• Original: https://github.com/facebookresearch/segment-anything
• Pretrain Dataset: SA - 1B

📦 Installation

No specific installation steps are provided in the original document.

💻 Usage Examples

Basic Usage

Image Classification

from urllib.request import urlopen
from PIL import Image
import timm

img = Image.open(urlopen(
    'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))

model = timm.create_model('samvit_large_patch16.sa1b', pretrained=True)
model = model.eval()

# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)

output = model(transforms(img).unsqueeze(0))  # unsqueeze single image into batch of 1

top5_probabilities, top5_class_indices = torch.topk(output.softmax(dim=1) * 100, k=5)

Image Embeddings

from urllib.request import urlopen
from PIL import Image
import timm

img = Image.open(urlopen(
    'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))

model = timm.create_model(
    'samvit_large_patch16.sa1b',
    pretrained=True,
    num_classes=0,  # remove classifier nn.Linear
)
model = model.eval()

# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)

output = model(transforms(img).unsqueeze(0))  # output is (batch_size, num_features) shaped tensor

# or equivalently (without needing to set num_classes=0)

output = model.forward_features(transforms(img).unsqueeze(0))
# output is unpooled, a (1, 256, 64, 64) shaped tensor

output = model.forward_head(output, pre_logits=True)
# output is a (1, num_features) shaped tensor

📚 Documentation

Explore the dataset and runtime metrics of this model in timm model results.

📄 License

This project is licensed under the Apache - 2.0 license.

📖 Citation

@article{kirillov2023segany,
  title={Segment Anything},
  author={Kirillov, Alexander and Mintun, Eric and Ravi, Nikhila and Mao, Hanzi and Rolland, Chloe and Gustafson, Laura and Xiao, Tete and Whitehead, Spencer and Berg, Alexander C. and Lo, Wan-Yen and Doll{'a}r, Piotr and Girshick, Ross},
  journal={arXiv:2304.02643},
  year={2023}
}

@article{dosovitskiy2020vit,
  title={An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale},
  author={Dosovitskiy, Alexey and Beyer, Lucas and Kolesnikov, Alexander and Weissenborn, Dirk and Zhai, Xiaohua and Unterthiner, Thomas and  Dehghani, Mostafa and Minderer, Matthias and Heigold, Georg and Gelly, Sylvain and Uszkoreit, Jakob and Houlsby, Neil},
  journal={ICLR},
  year={2021}
}

@misc{rw2019timm,
  author = {Ross Wightman},
  title = {PyTorch Image Models},
  year = {2019},
  publisher = {GitHub},
  journal = {GitHub repository},
  doi = {10.5281/zenodo.4414861},
  howpublished = {\url{https://github.com/huggingface/pytorch-image-models}}
}