Dataset Viewer
Duplicate
The dataset viewer is not available for this split.
Server error while post-processing the rows. Please report the issue.
Error code:   RowsPostProcessingError

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

HLS v2 change-detection point series

Point time series sampled from NASA HLS v2 for two change-detection benchmarks. One compressed NumPy archive per task. Inside each archive, one .npz per MGRS tile and year.

Task Points Tile-years Years
Wildfire 840 1565 2020–2024, every point, every year
Hansen forest loss 1135 4764 2020–2025, every point, every year

Archives: wildfire.tar, hansen.tar.

Observation product

Each .npz keeps every HLS granule that has a real reflectance value at the point. Cloudy observations stay in. Reflectance is the original int16 digital number (reflectance = DN * 0.0001). Fill is -9999. Fmask is the raw byte; fill is 255.

Bands, in order: B01, B02, B03, B04, B05, B06, B07, B08, B8A, B09, B10, B11, B12. Landsat-30 has no red-edge or B08/B8A/B12 slot, so those positions are -9999. Sensor code: 1 = S30, 2 = L30.

Arrays:

  • point_index, lon, lat, label, sample_id: the points in this tile-year
  • obs_point: row into point_index for each observation
  • obs_time: yyyyDDDTHHmmss
  • obs_sensor, fmask, data with shape (observations, 13)
  • bands, year, tile

A tile-year with no usable observation is still present, with zero rows in data.

Wildfire label is the class id (0 or 1) and sample_id is the plot id. Hansen label is loss_year_actual and sample_id is the site id. Change dates stay in the source tables used to choose the points.

Downloads last month
66