Update README.md

README.md

@@ -13,6 +13,47 @@ configs:
 - config_name: unlabeled
   data_files: "unlb/*.csv"
   description: "Unlabeled spectra data for semi-supervised or self-supervised learning."
+---
+# Spectra Dataset for Trait Prediction
+
+This dataset includes hyperspectral reflectance data of plant samples. It is intended for machine learning tasks such as Regression for plant trait prediction with self and semi-supervised learning.
+
+## 📁 Configurations
+
+### 1. `labeled`
+- File: `lb.csv`
+- Contains:
+  - Sample ID
+  - Spectral bands (400-2450 nm) >> 1721 bands
+  - Trait measurements (e.g., nitrogen, moisture, etc.)
+
+| Column         | Description                                  |
+|----------------|----------------------------------------------|
+| `dataset_id`    | Unique sample identifier                     |
+| `400` | Reflectance at 400nm                         |
+| ...            | More spectral bands (every 10nm, etc.)       |
+| `2450`         | More spectral bands (every 10nm, etc.)       |
+| Cp             | Nitrogen content                     |
+| Cm             |                        |
+| Cw             |                            |
+| LAI            |                          |
+| Cab            |                           |
+| Car            |                          |
+| Anth           |                           |
+---
+
+### 2. `unlabeled`
+- Files: all CSVs under `Unlb/`
+- Contains:
+  - Sample ID
+  - Spectral bands only
+
+| Column         | Description                                  |
+|----------------|----------------------------------------------|
+| `dataset_id`   | Unique sample identifier                     |
+| `400` | Reflectance at 400nm                         |
+|            |                           |
+---
 
 license: mit
 ---