Skip to content

Workshop: General information

Jump to bottom
kais-siala edited this page Nov 26, 2019 · 4 revisions

General information

Organization

  • Duration: 3 hours
  • Number of participants: up to 8, plus organizer
  • Room equipment: White board or flipchart, TV screen or projector

Agenda

  • T0 Introduction and expectations (10 min)
  • T1 Workflow of the clustering algorithm - paper version (30 min) including explanation for k-means and max-p
  • T2 Explanation of new features and user settings (20 min)
  • Break (10 min)
  • T3 Applications on Ghana (40 min)
  • T4 Comparison and discussion of results (20 min)
  • T5 Clustering of transmission lines (15 min)
  • T6 Open questions (10 min)

Preparation (participants)

It would be nice if you could skim through this paper (at least have a look at all the figures): https://doi.org/10.1016/j.esr.2019.100362

For an active participation, I recommend that you...

  • install conda, choose a directory where you would like to have the repository and start a command promt there
  • clone the repository on your computer:

$ ~\directory-of-my-choice> git clone https://github.com/tum-ens/geoclustering.git

  • install the environment (warning: eventually edit the prefix to point to your conda envs folder)

$ cd geoclustering\env\

$ conda env create -f geoclustering.yml

$ conda activate geoclustering

  • download the folder with the examples to be used in the workshop and put it in the same folder as the repository (eventually replace an existing one):

https://syncandshare.lrz.de/getlink/fiHYsEHxgJBhM342MJhQKg3E/

You can test whether everything is fine by going through these steps:

~\directory-of-my-choice\geoclustering> git checkout Example_Ghana

$ conda activate geoclustering

$ cd code

$ python runme.py

Software recommendations: QGIS, Notepad++

If you encounter any issues, please contact Kais Siala [email protected]. You can also attend the workshop without installing the software, but you will not be able to run the experiments on your computer.

Preparation (organizer)

  • Prepare examples (e.g. Ghana): run the scripts for these countries and compress the output in one RAR file / send a link to a shared folder.
  • Create a separate branch for each example, with the corresponding settings in config.py and runme.py.
Clone this wiki locally