Downloads · 30 days
0
maritime-vessel-recognition-ssl-finetuned
hanchong/maritime-vessel-recognition-ssl-finetuned
maritime-vessel-recognition-ssl-finetuned is a image classification model from hanchong. Use it when you need a label for an image.
The models shown below were fine-tuned end-to-end on the in-distribution maritime vessel dataset starting from SSL-pretrained ViT-L-16 backbones.
Access
Public
Updated May 13, 2026
Repo size
38.9 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pth38.9 GB · 100%
At a glance
- Task
- Image Classification
- Access
- Public
- Created
- May 12, 2026
- Updated
- May 13, 2026
- SHA
- 963ced2c
Datasets
Model card
From the Hugging Face model README
Maritime Vessel Recognition (Fine-Tuned)
The models shown below were fine-tuned end-to-end on the in-distribution maritime vessel dataset starting from SSL-pretrained ViT-L-16 backbones.
Models
| Checkpoint Name | Loss Function | Balanced Training Data | SSL Backbone |
|---|---|---|---|
| vit-l-16-franca.pth | Cross Entropy | - | Franca |
| vit-l-16-franca_BTD.pth | Cross Entropy | 200k per category | Franca |
| vit-l-16-franca_LAL.pth | Logit Adjusted Loss | - | Franca |
| vit-l-16-franca_BTDandLAL.pth | Logit Adjusted Loss | 200k per category | Franca |
| vit-l-16-simdinov2.pth | Cross Entropy | - | SimDINOv2 |
| vit-l-16-simdinov2_BTD.pth | Cross Entropy | 200k per category | SimDINOv2 |
| vit-l-16-simdinov2_LAL.pth | Logit Adjusted Loss | - | SimDINOv2 |
| vit-l-16-simdinov2_BTDandLAL.pth | Logit Adjusted Loss | 200k per category | SimDINOv2 |
Evaluation
- Task
- Image Classification
- Created
- May 12, 2026
- Updated
- May 13, 2026