Lets-Plot Kotlin API

A Grammar of Graphics for Kotlin.

Lets-Plot Kotlin API is a Kotlin API for Lets-Plot Multiplatform plotting library, \ which is built on the principles of layered graphics first described in the \ Leland Wilkinson work The Grammar of Graphics.

Lets-Plot Kotlin API is largely based on the API
provided by ggplot2 package well-known to data scientists who use R.

To learn more about the Grammar of Graphics,
we recommend an excellent book called
“ggplot2: Elegant Graphics for Data Analysis”.

This will be a good prerequisite for further exploration of the Lets-Plot library.

Quickstart

Inside Kotlin Notebook, Datalore or Jupyter with Kotlin Kernel:

%use lets-plot

val rand = java.util.Random()
val data = mapOf(
    "rating" to List(200) { rand.nextGaussian() } + List(200) { rand.nextGaussian() * 1.5 + 1.5 },
    "cond" to List(200) { "A" } + List(200) { "B" }
)

var p = letsPlot(data)
p += geomDensity(color = "dark_green", alpha = .3) { x = "rating"; fill = "cond" }
p + ggsize(700, 350)

See the "Quickstart" notebook in Datalore or Jupyter nbviewer.

Usage
Documentation
What is new in 4.8.0
Recent Updates in the Gallery
Change Log
Code of Conduct
License

Usage

Notebooks

With the help of Lets-Plot Kotlin API you can easily create plots in Kotlin Notebook, Datalore, Jupyter with Kotlin Kernel \ or any other notebook that supports Kotlin Kernel.

"Line Magics"

%use lets-plot

This "line magic" will apply Lets-Plot library descriptor which adds to your notebook all the boilerplate code necessary to create plots.

By default, library descriptor is bundled with the Kotlin Jupyter Kernel installed in your environment. \ However, you can override the default settings using:

%useLatestDescriptors

In this case the latest library descriptor will be pulled from the Kotlin Jupyter Libraries repository.

Library Descriptor Parameters

%use lets-plot(api=4.8.0, lib=4.4.1, js=4.4.1, isolatedFrame=false)

api - version of the Lets-Plot Kotlin API.
lib - version of the Lets-Plot Multiplatform (JARs).
js - version of the Lets-PLot Multiplatform JavaScript bundle.
isolatedFrame - If false: load JS just once per notebook (default in Jupyter). If true: include Lets-Plot JS in each output (default in Datalore notebooks).

Compose Multiplatform

To learn how to embed Lets-Plot charts in Compose Multiplatform applications, please check out the Lets-Plot Skia Frontend project.

JVM and Kotlin/JS

To learn more about creating plots in JVM or Kotlin/JS environment please read USAGE_SWING_JFX_JS.md.

Examples

Examples of using of the Lets-Plot Kotlin API in JVM and Kotlin/JS applications are available in the Lets-Plot Kotlin Mini Apps (Demos) GitHub repository.

Documentation

Lets-Plot Kotlin API documentation and API reference: Lets-Plot for Kotlin
A quick introduction to the Grammar of Graphics and Lets-Plot Kotlin API: Lets-Plot Usage Guide

What is new in 4.8.0

Waterfall Plot

See example notebook.
geomBand():

See example notebook.
Custom Legends
- manualKey parameter in plot layer
- layerKey() function
See example notebook.
Customizing Legends Appearance

Override aesthetics in the guideLegend() function.

See example notebook.
And More

See CHANGELOG.md for a full list of changes.