Model card for levit_conv_128.fb_dist_in1k

A LeViT image classification model using default linear mode (non-convolutional mode with nn.Linear and nn.BatchNorm1d). Pretrained on ImageNet-1k using distillation by paper authors.

Model Details

• Model Type: Image classification / feature backbone
• Model Stats:
  • Params (M): 9.2
  • GMACs: 0.4
  • Activations (M): 2.7
  • Image size: 224 x 224
• Papers:
  • LeViT: a Vision Transformer in ConvNet's Clothing for Faster Inference: https://arxiv.org/abs/2104.01136
• Original: https://github.com/facebookresearch/LeViT
• Dataset: ImageNet-1k

Model 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('levit_conv_128.fb_dist_in1k', 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(
    'levit_conv_128.fb_dist_in1k',
    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 (ie.e a (batch_size, num_features, H, W) tensor

output = model.forward_head(output, pre_logits=True)
# output is (batch_size, num_features) tensor

Feature Map Extraction

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(
    'levit_conv_128.fb_dist_in1k',
    pretrained=True,
    features_only=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

for o in output:
    # print shape of each feature map in output
    # e.g. for levit_conv_256: 
    # torch.Size([2, 256, 14, 14])
    # torch.Size([2, 384, 7, 7])
    # torch.Size([2, 512, 4, 4])
    print(o.shape)

Model Comparison

Citation

@InProceedings{Graham_2021_ICCV,
  author    = {Graham, Benjamin and El-Nouby, Alaaeldin and Touvron, Hugo and Stock, Pierre and Joulin, Armand and Jegou, Herve and Douze, Matthijs},
  title     = {LeViT: A Vision Transformer in ConvNet's Clothing for Faster Inference},
  booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
  month     = {October},
  year      = {2021},
  pages     = {12259-12269}
}

@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/rwightman/pytorch-image-models}}
}