regnety_640.seer_ft_in1k

Model card for regnety_640.seer_ft_in1k

A RegNetY-64GF image classification model. Pretrained according to SEER: self-supervised learning with SwAV on "2B random internet images". Fine-tuned on ImageNet-1k by paper authors.

SEER is licensed under SEER license, Copyright (c) Meta Platforms, Inc. All Rights Reserved. The license is a non-commercial license with useage and distribution restrictions.

The timm RegNet implementation includes a number of enhancements not present in other implementations, including:

• stochastic depth
• gradient checkpointing
• layer-wise LR decay
• configurable output stride (dilation)
• configurable activation and norm layers
• option for a pre-activation bottleneck block used in RegNetV variant
• only known RegNetZ model definitions with pretrained weights

Model Details

• Model Type: Image classification / feature backbone
• Model Stats:
  • Params (M): 281.4
  • GMACs: 188.5
  • Activations (M): 124.8
  • Image size: 384 x 384
• Papers:
  • Self-supervised Pretraining of Visual Features in the Wild: https://arxiv.org/abs/2103.01988v2
  • Designing Network Design Spaces: https://arxiv.org/abs/2003.13678
• Original: https://github.com/facebookresearch/vissl
• Dataset: ImageNet-1k
• Pretrain Dataset: RandomInternetImages-2B

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('regnety_640.seer_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(
    'regnety_640.seer_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, 192, 192])
    #  torch.Size([1, 328, 96, 96])
    #  torch.Size([1, 984, 48, 48])
    #  torch.Size([1, 1968, 24, 24])
    #  torch.Size([1, 4920, 12, 12])

    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(
    'regnety_640.seer_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, 4920, 12, 12) 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.

For the comparison summary below, the ra_in1k, ra3_in1k, ch_in1k, sw_, and lion_ tagged weights are trained in timm.

Citation

@article{goyal2022vision,
  title={Vision Models Are More Robust And Fair When Pretrained On Uncurated Images Without Supervision}, 
  author={Priya Goyal and Quentin Duval and Isaac Seessel and Mathilde Caron and Ishan Misra and Levent Sagun and Armand Joulin and Piotr Bojanowski},
  year={2022},
  eprint={2202.08360},
  archivePrefix={arXiv},
  primaryClass={cs.CV}
}

@InProceedings{Radosavovic2020,
  title = {Designing Network Design Spaces},
  author = {Ilija Radosavovic and Raj Prateek Kosaraju and Ross Girshick and Kaiming He and Piotr Doll{'a}r},
  booktitle = {CVPR},
  year = {2020}
}

@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}}
}