vpipkt / raster-frames

Spark DataFrames for earth observation data

^™

RasterFrames™ brings the power of Spark DataFrames to geospatial raster data, empowered by the map algebra and tile layer operations of GeoTrellis.

RasterFrames™ is a new project under active development. Feedback and contributions are welcomed as we continue to improve it.

Getting Started

Quick Start

macOS

1. If not already, install Homebrew
2. Run brew install sbt
3. Run sbt new s22s/raster-frames.g8

Linux

1. Install sbt
2. Run sbt new s22s/raster-frames.g8

Windows

1. Install sbt
2. Run sbt new s22s/raster-frames.g8

General Setup

RasterFrames is published via Bintray's JCenter server. In sbt you introduce it with this build setting:

resolvers += Resolver.jcenterRepo

(See this for use of JCenter with other build tools.)

To use RasterFrames, add the following library dependency:

sbt:

libraryDependencies += "io.astraea" %% "raster-frames" % "x.y.z"

Maven:

<dependency>
  <groupId>io.astraea</groupId>
  <artifactId>raster-frames</artifactId>
  <version>x.y.z</version>
</dependency>

Note: It assumes that SparkSQL 2.1.1 is available in the runtime classpath.

Documentation

• Users' Manual
• API Documentation
• List of available UDFs

Copyright and License

All code is available to you under the Apache 2.0 License, copyright Astraea, Inc. 2017.