Downloads · 30 days
0
IverMartinsen/scampi-dino-vits16
scampi-dino-vits16 is a machine learning model from IverMartinsen. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
This repo contains checkpoints and weights for ViT-S and ViT-B trained on 1.6M microfossils using DINO. These models were developed as foundation models for microfossil analysis and were presented in The Fossil Fronti…
Downloads · 30 days
0
Access
Public
Updated May 16, 2025
Repo size
2.9 GB
Likes
1
Public
Click a slice to open those files.
.pth2.9 GB · 100%
From the Hugging Face model README
This repo contains checkpoints and weights for ViT-S and ViT-B trained on 1.6M microfossils using DINO. These models were developed as foundation models for microfossil analysis and were presented in The Fossil Frontier: An answer to the 3 billion fossil question. Please visit our GitHub frontend for source code, evaluation benchmarks, and example of usage.
To download models go to 'files and versions', or simply use the links in this table:
<table> <tr> <th>arch</th> <th>params</th> <th>nn f1</th> <th>p@n_k</th> <th colspan="4">download</th> </tr> <tr> <td>ViT-S/16</td> <td>21M</td> <td>0.91</td> <td>0.60</td> <td><a href="https://huggingface.co/IverMartinsen/scampi-dino-vits16/resolve/main/vit_small_backbone.pth?download=true">backbone only</a></td> <td><a href="https://huggingface.co/IverMartinsen/scampi-dino-vits16/resolve/main/vit_small_checkpoint.pth?download=true">full ckpt</a></td> <td><a href="https://huggingface.co/IverMartinsen/scampi-dino-vits16/resolve/main/vit_small_args.txt?download=true">args</a></td> <td><a href="https://huggingface.co/IverMartinsen/scampi-dino-vits16/resolve/main/vit_small_log.txt?download=true">logs</a></td> </tr> <tr> <td>ViT-B/16</td> <td>85M</td> <td>0.90</td> <td>0.60</td> <td><a href="https://huggingface.co/IverMartinsen/scampi-dino-vits16/resolve/main/vit_base_backbone.pth?download=true">backbone only</a></td> <td><a href="https://huggingface.co/IverMartinsen/scampi-dino-vits16/resolve/main/vit_base_checkpoint.pth?download=true">full ckpt</a></td> <td><a href="https://huggingface.co/IverMartinsen/scampi-dino-vits16/resolve/main/vit_base_args.txt?download=true">args</a></td> <td><a href="https://huggingface.co/IverMartinsen/scampi-dino-vits16/resolve/main/vit_base_log.txt?download=true">logs</a></td> </tr> </table>These models were trained on the LUMI supercomputer using a single node of eight GPUs (four AMD MI250X GPUs). We acknowledge Sigma2, Norway for awarding this project access to the LUMI supercomputer, owned by the EuroHPC Joint Undertaking, hosted by CSC (Finland) and the LUMI consortium through Sigma2, Norway.
Please consider citing the Fossil Frontier paper if using these models.
BibTeX:
[More Information Needed]
APA:
[More Information Needed]