Model card for unireplknet_l.in22k_ft_in1k_384

A UniRepLKNet image classification model. Trained on ImageNet-21k by the paper authors.

UniRepLKNet is a large-kernel ConvNet that uses Dilated Reparameterization Blocks: at training time the depthwise convolution is materialized as a sum of parallel small-kernel dilated branches; at inference these branches fuse into a single equivalent large-kernel depthwise conv.

Model Details

• Model Type: Image classification / feature backbone
• Model Stats:
  • Params (M): 220.08
  • GMACs: 106.36
  • Image size: 384 x 384
• Papers:
  • UniRepLKNet, A Universal Perception Large-Kernel ConvNet for Audio, Video, Point Cloud, Time-Series and Image Recognition: https://arxiv.org/abs/2311.15599
• Original: https://github.com/AILab-CVC/UniRepLKNet
• Pretrain Dataset: ImageNet-21k

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('unireplknet_l.in22k_ft_in1k_384', pretrained=True)
model = model.eval()

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))

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

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(
    'unireplknet_l.in22k_ft_in1k_384',
    pretrained=True,
    features_only=True,
)
model = model.eval()

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))

for o in output:
    print(o.shape)

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(
    'unireplknet_l.in22k_ft_in1k_384',
    pretrained=True,
    num_classes=0,
)
model = model.eval()

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 = model.forward_features(transforms(img).unsqueeze(0))
output = model.forward_head(output, pre_logits=True)

Citation

@article{ding2023unireplknet,
  title={UniRepLKNet: A Universal Perception Large-Kernel ConvNet for Audio, Video, Point Cloud, Time-Series and Image Recognition},
  author={Ding, Xiaohan and Zhang, Yiyuan and Ge, Yixiao and Zhao, Sijie and Song, Lin and Yue, Xiangyu and Shan, Ying},
  journal={arXiv preprint arXiv:2311.15599},
  year={2023}
}

@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}}
}