Image Segmentation
fastai
PyTorch
geospatial
remote-sensing
semantic-segmentation
sentinel-1
alphaearth
surface-water
Instructions to use rohitm9/surfaceWaterGlobal with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- fastai
How to use rohitm9/surfaceWaterGlobal with fastai:
from huggingface_hub import from_pretrained_fastai learn = from_pretrained_fastai("rohitm9/surfaceWaterGlobal") - Notebooks
- Google Colab
- Kaggle
Download MODEL_INFO.json from rohitm9/surfaceWaterGlobal: direct link, hf CLI and curl.
- Browser
- Download file 2.37 kB
-
https://huggingface.co/rohitm9/surfaceWaterGlobal/resolve/main/MODEL_INFO.json
- Command line
-
hf download hf://rohitm9/surfaceWaterGlobal/MODEL_INFO.json
-
curl -L -o MODEL_INFO.json https://huggingface.co/rohitm9/surfaceWaterGlobal/resolve/main/MODEL_INFO.json
2.37 kB
| { | |
| "model_id": "rohitm9/surfaceWaterGlobal", | |
| "version": "2.0.0-labelclass", | |
| "framework": [ | |
| "pytorch", | |
| "fastai" | |
| ], | |
| "task": "binary-semantic-segmentation", | |
| "architecture": "resnet34-unet-with-aef-bottleneck", | |
| "training_target": "Dynamic World label band, class 0 (water)", | |
| "training_tiles": { | |
| "total": 5278, | |
| "train": 4222, | |
| "valid": 1056 | |
| }, | |
| "inputs": { | |
| "sentinel_1": { | |
| "bands": 3, | |
| "order": [ | |
| "VV", | |
| "VH", | |
| "angle" | |
| ], | |
| "source": "Earth Engine COPERNICUS/S1_GRD, dB, 10 m" | |
| }, | |
| "alphaearth": { | |
| "bands": 64, | |
| "order": "source order", | |
| "source": "AlphaEarth Foundations annual embeddings v1" | |
| } | |
| }, | |
| "aef_projection": { | |
| "input_channels": 64, | |
| "output_channels": 16, | |
| "kernel_size": 1 | |
| }, | |
| "classes": [ | |
| "other", | |
| "water" | |
| ], | |
| "checkpoint": "models/s1aef_bottleneck_resnet34_best.pth", | |
| "normalization": "band_stats.npz", | |
| "checkpoint_sha256": "dd0252e2d08c693e3f88949a0a34a846f91dc7ab1668d91ba7ac0ddb688b06ec", | |
| "thresholds": { | |
| "validation_selected": 0.5, | |
| "paper_reference_set": 0.3, | |
| "test_time_augmentation": "four flips" | |
| }, | |
| "variants": { | |
| "variants/s1_only_k0_seed42": { | |
| "model_kind": "s1", | |
| "checkpoint": "models/s1dw_resnet34_best.pth", | |
| "checkpoint_sha256": "a5eab4f07d68256d5a7922a8943efb1b0a14cbc0a513315aff737426a57b6e68" | |
| }, | |
| "variants/s1_aef_previous_year_k16_seed42": { | |
| "model_kind": "s1aef", | |
| "checkpoint": "models/s1aef_bottleneck_resnet34_best.pth", | |
| "checkpoint_sha256": "5bd312409fcf86e0a49258c37aa95a53867a5a6b2cef4f73d2e527a3d6fb2717" | |
| } | |
| }, | |
| "evaluation_status": "published", | |
| "evaluation": { | |
| "validation_split": { | |
| "water_iou": 0.697666, | |
| "threshold": 0.5, | |
| "tta": true | |
| }, | |
| "planetscope_53_scenes": { | |
| "threshold": 0.3, | |
| "pooled_water_iou": 0.850797, | |
| "per_scene_mean_water_iou": 0.755753 | |
| }, | |
| "s1s2water_15_scenes": { | |
| "grid_m": 30, | |
| "threshold": 0.5, | |
| "pooled_water_iou_mean_3_seeds": 0.9529, | |
| "opera_dswx_s1_pooled_water_iou": 0.869 | |
| } | |
| }, | |
| "supersedes": { | |
| "tag": "v1-mixed-labels", | |
| "checkpoint_sha256": "3b89c5b768be8d8b5a00260693ca5d2a051ec63969234702903677b9e326138f", | |
| "reason": "trained on a mixture of binarized DW water probability and DW label class" | |
| } | |
| } | |