rexnetr_200.sw_in12k_ft_in1k

Model card for rexnetr_200.sw_in12k_ft_in1k

A ReXNet-R image classification model. The R variant of the architecture is timm specific and rounds channels (modulus 8 or 16) to prevent performance issues w/ NVIDIA Tensor Cores. Pretrained on ImageNet-12k and fine-tuned on ImageNet-1k by Ross Wightman in timm.

Model Details

• Model Type: Image classification / feature backbone
• Model Stats:
  • Params (M): 16.5
  • GMACs: 1.6
  • Activations (M): 15.1
  • Image size: train = 224 x 224, test = 288 x 288
• Papers:
  • Rethinking Channel Dimensions for Efficient Model Design: https://arxiv.org/abs/2007.00992
• Original: https://github.com/huggingface/pytorch-image-models
• Dataset: ImageNet-1k
• Pretrain Dataset: ImageNet-12k

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('rexnetr_200.sw_in12k_ft_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)

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(
    'rexnetr_200.sw_in12k_ft_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.:
    #  torch.Size([1, 32, 112, 112])
    #  torch.Size([1, 80, 56, 56])
    #  torch.Size([1, 120, 28, 28])
    #  torch.Size([1, 256, 14, 14])
    #  torch.Size([1, 368, 7, 7])

    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(
    'rexnetr_200.sw_in12k_ft_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, a (1, 2560, 7, 7) shaped tensor

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

Model Comparison

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

Citation

@misc{han2021rethinking,
  title={Rethinking Channel Dimensions for Efficient Model Design}, 
  author={Dongyoon Han and Sangdoo Yun and Byeongho Heo and YoungJoon Yoo},
  year={2021},
  eprint={2007.00992},
  archivePrefix={arXiv},
  primaryClass={cs.CV}
}  

@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}}
}