Update README.md
Browse files
README.md
CHANGED
|
@@ -44,6 +44,8 @@ size_categories:
|
|
| 44 |
|
| 45 |
# Dist_Uncertainty – Hyperspectral Case Studies for Plant Trait Uncertainty Assessment
|
| 46 |
|
|
|
|
|
|
|
| 47 |
## Dataset Description
|
| 48 |
|
| 49 |
This dataset contains hyperspectral remote sensing imagery and associated land cover labels used as **out-of-domain (OOD) test cases** for evaluating uncertainty estimation methods in deep learning-based plant trait retrievals. It accompanies the paper by Cherif et al. (2025, *Biogeosciences*) and supports the evaluation of a distance-based uncertainty method (Dis_UN) against traditional approaches (deep ensembles and Monte Carlo dropout).
|
|
@@ -55,6 +57,8 @@ The dataset includes two real-world hyperspectral scenes from different sensors
|
|
| 55 |
|
| 56 |
Each scene comes with spectral band metadata and pixel-level land cover labels, enabling the identification of OOD components such as urban surfaces, bare ground, water bodies, and clouds — elements not represented in the training data.
|
| 57 |
|
|
|
|
|
|
|
| 58 |
The `TrainingSpectra_1522.csv` file serves as the **reference distribution** for the distance-based uncertainty method (Dis_UN): at inference time, the dissimilarity between any new unseen spectrum and this training set is computed in the predictor and embedding space to quantify how far the new data lies from the known training distribution.
|
| 59 |
|
| 60 |
---
|
|
|
|
| 44 |
|
| 45 |
# Dist_Uncertainty – Hyperspectral Case Studies for Plant Trait Uncertainty Assessment
|
| 46 |
|
| 47 |
+
|
| 48 |
+
|
| 49 |
## Dataset Description
|
| 50 |
|
| 51 |
This dataset contains hyperspectral remote sensing imagery and associated land cover labels used as **out-of-domain (OOD) test cases** for evaluating uncertainty estimation methods in deep learning-based plant trait retrievals. It accompanies the paper by Cherif et al. (2025, *Biogeosciences*) and supports the evaluation of a distance-based uncertainty method (Dis_UN) against traditional approaches (deep ensembles and Monte Carlo dropout).
|
|
|
|
| 57 |
|
| 58 |
Each scene comes with spectral band metadata and pixel-level land cover labels, enabling the identification of OOD components such as urban surfaces, bare ground, water bodies, and clouds — elements not represented in the training data.
|
| 59 |
|
| 60 |
+
|
| 61 |
+
|
| 62 |
The `TrainingSpectra_1522.csv` file serves as the **reference distribution** for the distance-based uncertainty method (Dis_UN): at inference time, the dissimilarity between any new unseen spectrum and this training set is computed in the predictor and embedding space to quantify how far the new data lies from the known training distribution.
|
| 63 |
|
| 64 |
---
|