Dataset · tanganke
Satellite imagery for land cover classification
tanganke/eurosat
A set of RGB images of the Earth's surface, labeled into 10 landscape and built-up categories.
What's inside
The dataset contains between 10,000 and 100,000 RGB satellite images. Each image is assigned a label from one of 10 classes: forests, pastures, annual crops, shrubland, roads, industrial or commercial buildings, and other terrain types.
What you can do with it
- Training satellite and aerial image classification models
- Benchmarking computer vision on remote sensing tasks
- Tasks
- Image classification
- Languages
- English
- Size
- 10K–100K rows
- Downloads this month
- 19.8K
- Likes
- 3
Models trained on it
- tanganke/clip-vit-base-patch32_eurosatEmbeddings 3.4K
- tanganke/clip-vit-large-patch14_eurosatEmbeddings 743
- tanganke/clip-vit-base-patch32_TA8_mtl 101
- tanganke/clip-vit-large-patch14_TA8_mtl 22
- AmeenAhmed2/zicoZero shot classification 0
- theblackcat102/clip-vit-base-patch16-task-experts-ep10Image classification 0
- theblackcat102/clip-vit-large-patch14-task-experts-ep10Image classification 0
How to load it
from datasets import load_dataset
ds = load_dataset("tanganke/eurosat")