BYOBNet (GENet, RepVGG, MobileOne, …)

Category: CNN · Code: timm/models/byobnet.py · Architectures: 53 · Pretrained weights: 57 (41 with ImageNet-1k results)

Bring-Your-Own-Blocks Network, a flexible network built from dataclass configs that stack blocks into stages. It implements GPU-Efficient Networks (gernet_*), RepVGG, MobileOne, RegNetZ, the ResNet image towers of CLIP (*_clip) and a number of ResNet style timm variants.

Papers

• Neural Architecture Design for GPU-Efficient Networks
• RepVGG: Making VGG-style ConvNets Great Again
• MobileOne: An Improved One millisecond Mobile Backbone
• Fast and Accurate Model Scaling
• Evolving Normalization-Activation Layers
• Learning Transferable Visual Models From Natural Language Supervision

Usage

See the Quickstart for loading pretrained weights, image preprocessing and fine-tuning, and Feature Extraction for using the models as backbones.

>>> import timm
>>> model = timm.create_model('repvgg_a2.rvgg_in1k', pretrained=True)
>>> data_config = timm.data.resolve_model_data_config(model)

Model API support

✓ supported, ✓ (square) works for square sizes only, ✗ not supported, – not needed (the model already accepts any input size). Checked by running each feature on a randomly initialized model (ByobNet with mobileone_s1 at 224), other architectures of the same class are expected to behave the same. See the Model API page for details.

Architectures

GMACs and parameter counts are for the default input size and classifier of each architecture, from the benchmark results.

Pretrained weights

Top-1 is ImageNet-1k validation accuracy from results-imagenet.csv at the listed eval size, - if the weight has no ImageNet-1k result (e.g. pretraining or image-text weights). Classes is the size of the pretrained classifier, none for weights without one.

Weights are grouped by source: the paper or origin given in the pretrained config, else a Hub organization other than timm, else the first part of the pretrained tag (architecture.tag). Sources with a single weight are listed under other.

rvgg tags

openai tags

ra2 tags

ch tags

apple tags

ra3 tags

yfcc15m tags

idstcv tags

cc12m tags

Other