The dataset for this challenge consists of satellite images collected from the Landsat series. The data has been pre-processed and divided into training and testing sets:

• Training: 1.1 million rows
• Test: 120,000 rows

The data has been labeled by in-situ field operations. Each row of the data corresponds to a 16x16 pixel patch extracted from a Landsat image. Each pixel patch has 6 spectral bands:

• Blue
• Green
• Red
• Near-infrared (NIR)
• Shortwave infrared 1 (SWIR1)
• Shortwave infrared 2 (SWIR2)

This dataset provides a rich and diverse source of information for participants to develop their models. It is important to note that the data may contain some level of noise or uncertainty due to the inherent limitations of satellite imagery and the labeling process.

Data available here.