Downloads · 30 days
0
geonusaf-unetformer-r18-random-fold2
sugam24/geonusaf-unetformer-r18-random-fold2
geonusaf-unetformer-r18-random-fold2 is a machine learning model from sugam24. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for pytorch.
Kathmandu Valley land-use segmentation, 6 classes, ignoreindex=255.
Access
Public
Updated Aug 17, 2026
Repo size
2.8 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt188 MB · 98%
At a glance
- Library
- pytorch
- Access
- Public
- Created
- Aug 17, 2026
- Updated
- Aug 17, 2026
- SHA
- 045a7ace
Model card
From the Hugging Face model README
GeoNUSAF - UNetFormer (ResNet-18) - random split, fold 2
Kathmandu Valley land-use segmentation, 6 classes, ignore_index=255.
| field | value |
|---|---|
| architecture | UNetFormer, timm resnet18 encoder (ImageNet), global-local attention decoder |
| split mode | random |
| fold | 2 of 3 |
| seed | 42 |
| input | 512x512, ImageNet norm, effective GSD 0.586 m/px |
| classes | Residential, Road, River, Forest, UnusedLand, Agricultural |
| lr (dec/enc) | 0.0006 / 6e-05, AdamW wd 0.0001 |
| aux head weight | 0.4 |
| best epoch | 98 |
| val mIoU | 0.5259 |
| val mF1 | 0.6726 |
| val OA | 0.7830 |
| val kappa | 0.6558 |
Per-class (validation)
| class | IoU | F1 |
|---|---|---|
| Residential | 0.8150 | 0.8980 |
| Road | 0.4114 | 0.5829 |
| River | 0.4927 | 0.6601 |
| Forest | 0.6710 | 0.8031 |
| UnusedLand | 0.2818 | 0.4397 |
| Agricultural | 0.4834 | 0.6517 |
Split caveat: block is sequence-block CV using an export-order proxy, not spatial CV.
Architecture is an independent implementation of Wang et al. (2022), ISPRS J. Photogramm. Remote Sens. 190:196-214 (the reference repo is GPL-3.0).
Checkpoint best.pt holds model_state plus cfg and metrics.
- Library
- pytorch
- Created
- Aug 17, 2026
- Updated
- Aug 17, 2026