Moonshine is a Python library that makes it easy for remote sensing researchers, professionals, and enthusiasts to develop ML models on their data. It provides pre-trained models across a variety of datasets and architectures, allowing you to reduce your labeling costs and compute requirements for your own application.

Why Use Moonshine?#

  1. Pretrained on multispectral data: Many existing packages are pretrained with ImageNet or similar RGB images. Using Moonshine you can unlock the full power of satellites that many contain many channels of multispectral data.

  2. Pretrained on remote sensing data: Pretraining in the domain of your data is important, and most off the shelf pretrained models are fit to natural images such as ImageNet.

  3. Focus on usability: While there are some academic remote sensing pretrained models available, they often are difficult to use and lack support. Moonshine is designed to be easy to use and will offer community support via Github and Slack.

Need more convincing that Moonshine works? Check out this comparison of Moonshine pretrained weights vs training from scratch:

Pretrain your models to save time and compute | width=400

The above chart shows the difference between training the functional map of the world classification task using our pre-trained model vs. training from scratch. The task is to classify patches of satellite data by the functional purpose of the land, with 63 possible classes and over 300,000 training images.

Training from scratch both performs worse overall, and for roughly the same level of accuracy we can train for 45% less time (28h vs 16h on a V100). Check out the quick start section for further information, including how to install the library.