birder-project/vit_reg1_s14_ls_dino-v2-dist-bio

image-feature-extraction
Actively maintainedby birder-project1913updated 2 days ago

vitreg1s14lsdino-v2-dist-bio is a compact Bio-DINO image encoder distilled from the larger Bio-DINO SoViT-150M/14 model. It keeps the same natural-photography biodiversity scope as the teacher model, but uses a much smaller ViT-S/14-style student with 21.7M backbone parameters and 384-dimensional…

README

tags: image-feature-extraction birder pytorch biology libraryname: birder license: apache-2.0 basemodel: birder-project/vitreg4so150mp14lsdino-v2-bio Model Card for vitreg1s14lsdino-v2-dist-bio vitreg1s14lsdino-v2-dist-bio is a compact Bio-DINO image encoder distilled from the larger Bio-DINO SoViT-150M/14 model. It keeps the same natural-photography biodiversity scope as the teacher model, but uses a much smaller ViT-S/14-style student with 21.7M backbone parameters and 384-dimensional…

Source attribution

  • HuggingFacebirder-project/vit_reg1_s14_ls_dino-v2-dist-bio

Related resources

vitreg4so150mp14ls_dino-v2-bio is a Bio-DINO image encoder for natural photographs of living organisms. It uses a SoViT-150M/14 Vision Transformer with 4 register tokens and 133.6M backbone parameters, trained with a DINOv2-style self-supervised objective on approximately 31 million curated images…

3.9K2 days ago

This repository contains the full Bio-DINO DINOv2 training weights for a SoViT-150M/14 Vision Transformer trained on natural photographs of living organisms. It is the companion release to the Birder backbone checkpoints at .

1324 days ago

# InstaNovo: De novo Peptide Sequencing Model ## Model Description

362 weeks ago
1.9K10 months ago
Python
48410 months ago
Python