CurveClassification

Minimal web application to train and evaluate random forests for classification of omics-type experiments.

Usage

A detailed description can be found in the provided PDF, but we also provide a video tutorial on Youtube

Building your own CurveClassification instance

Requirenment

• Docker

  • Installation on linux
  • Installation on Windows 10
  • Installation on MacOS

  since June 20 you have to register to be able to download docker for Windows and MacOS.

Execution

Hint: The first installation of the programm takes up to 15 minutes.

Linux

You need sudo rights if your user is not part of the group docker

git clone https://github.com/kusterlab/curveClassification_shiny.git
make run

Windows

1. Open PowerShell
2. docker build https://github.com/kusterlab/curveClassification_shiny -t curveclassification
3. docker run -p 8787:8787 curveclassification

Step 2 is not necessary after the first run.
*Hint*: docker on Windows reacts with huge delays.

#### MacOS
You can follow the installation for Linux


The server/computer should now run a *CurveClassification* instance, which is available on port `8787`.