README.md

---
license: cc-by-nc-4.0
---


# 🌲 AGBD: A Global-scale Biomass Dataset 🌳

Authors: Ghjulia Sialelli ([[email protected]](mailto:[email protected])), Torben Peters, Jan Wegner, Konrad Schindler



## 🚀 Usage

To get started with this dataset, use the following code snippet:

```python
# Install the datasets library if you haven't already
!pip install datasets

# Import necessary modules
from datasets import load_dataset

# Load the dataset
dataset = load_dataset('prs-eth/AGBD', trust_remote_code=True, streaming=True)["train"] # Options: "train", "val", "test"
```

This code will load the dataset as an `IterableDataset`. You can find more information on how to work with `IterableDataset` objects in the [Hugging Face documentation](https://huggingface.co/docs/datasets/access#iterabledataset).

---


## 📊 Dataset Overview

Each sample in the dataset contains a **pair of pre-cropped, pre-normalized images** along with their corresponding **biomass labels**. For additional resources, including links to the preprocessed uncropped data, please visit the [project page on GitHub](https://github.com/ghjuliasialelli/AGBD/).

### ⚙️ Load Dataset Options

The `load_dataset` function provides the following configuration options:

- **`normalize_data`**: `{True, False}` 
 Whether to return **normalized (0-1) data** or **raw values**.

- **`additional_features`**: `[]` 
 A list of additional features the dataset should include. *Refer to the documentation below for more details.*

- **`patch_size`**: `25` 
 The size of the returned patch (in pixels). The maximum value is **25 pixels**, which corresponds to **250 meters**.

---
### 🖼️ Image Details

Each image consists of **24 channels**, organized into the following categories:

- **Spectral Bands**: 
 `B01, B02, B03, B04, B05, B06, B07, B08, B8A, B09, B11, B12`

- **Geographical Coordinates**: 
 `lat_cos, lat_sin, lon_cos, lon_sin`

- **ALOS PALSAR Bands**: 
 `alos_hh, alos_hv`

- **Canopy Heights**: 
 `ch, ch_std`

- **Land Cover Information**: 
 `lc_cos, lc_sin, lc_prob`

- **Digital Elevation Model (DEM)**: 
 `dem`

---
### 📐 Channel Structure

The channels are structured as follows:

```plaintext
(Spectral Bands) | (Geographical Coordinates) | (ALOS PALSAR Bands) | (Canopy Heights) | (Land Cover Information) | DEM
```

```plaintext
(B01 B02 B03 B04 B05 B06 B07 B08 B8A B09 B11 B12) | (lat_cos, lat_sin, lon_cos, lon_sin) | (alos_hh, alos_hv) | (ch, ch_std) | (lc_cos, lc_sin, lc_prob) | dem
```
---
### ➕ Additional Features

You can include a list of additional features from the options below in your dataset configuration:
- **`"agbd_se"` - AGBD Standard Error**: The uncertainty estimate associated with the aboveground biomass density prediction for each GEDI footprint.
- **`"elev_lowes"` - Elevation**: The height above sea level at the location of the GEDI footprint.
- **`"leaf_off_f"` - Leaf-Off Flag**: Indicates whether the measurement was taken during the leaf-off season, which can impact canopy structure data.
- **`"pft_class"` - Plant Functional Type (PFT) Class**: Categorization of the vegetation type (e.g., deciduous broadleaf, evergreen needleleaf).
- **`"region_cla"` - Region Class**: The geographical area where the footprint is located (e.g., North America, South Asia).
- **`"rh98"` - RH98 (Relative Height at 98%)**: The height at which 98% of the returned laser energy is reflected, a key measure of canopy height.
- **`"sensitivity"` - Sensitivity**: The proportion of laser pulse energy reflected back to the sensor, providing insight into vegetation density and structure.
- **`"solar_elev"` - Solar Elevation**: The angle of the sun above the horizon at the time of measurement, which can affect data quality.
- **`"urban_prop"` - Urban Proportion**: The percentage of the footprint area that is urbanized, helping to filter or adjust biomass estimates in mixed landscapes.
- **`"gedi_num_days"` - Date of GEDI Footprints**: The specific date on which each GEDI footprint was captured, adding temporal context to the measurements.
- **`"s2_num_days"` - Date of Sentinel-2 Image**: The specific date on which each Sentinel-2 image was captured, ensuring temporal alignment with GEDI data.
- **`"lat"` - Latitude**: Latitude of the central pixel.
- **`"lon"` - Longitude**: Longitude of the central pixel.