Page tree

On the analysis itself: We're still working on some details of the data analysis. Since this concerns the initial data format, we will first fix this before we'll supply you with extensive examples. There will also be a bit of a process throughout the program, since it's hard to anticipate how fast the students will be on analysing their data. We will of course keep you updated with some code examples and details on what they should be able to get out of the analysis.

Fixed facts on the analysis:
 * We will use Jupyter notebooks. The DESY IT runs a hub that is running on the NAF servers.
 * The programming language is python+ROOT
 * The data will be supplied in the form of ROOT TTrees
 * The two teams will perform different experiments, thus it will be two different data sets and two analysis approaches.
 * We're in the act of compiling example notebooks for you. Those will contain one possible way to analyse the data, hopefully with a lot of helpful comments. Keep in mind that many roads lead to Rome and try to advise the students accordingly.

On the Jupyter notebooks:

  1. You can access the hub here: https://naf-jhub.desy.de/
  2. Log in with your DESY credentials
  3. Spawn a server
    1. Primary group: Default (or choose your group)
    2. GPU node: No
    3. Jupyter Launch Modus: Classical Notebook
    4. Hit "Spwan"
    5. Last year it got stuck here from time to time. Please just try again and report to me if this is the case.
  4. You are now in your afs home directory.
  5. To create a new notebook in the top right corner hit New→Python3. You'll also see C++ ROOT - I have not tested this yet and recommend for the students to work with python.
  6. Play around with it a bit, using python and ROOT syntax (don't forget to import ROOT)

I put a few explanations and examples into the notebook attached. You can open it by following until step 4 and then clicking "Upload" at the top right.

There is also one more tutorial on ROOT in Jupyter notebooks that you can check out: https://root.cern.ch/notebooks/HowTos/HowTo_ROOT-Notebooks.html


CERN SWAN

The CERN SWAN Service is very similar. Similar to the workflow mentioned above, follow these steps:

  1. Access the hub here: https://swan.cern.ch/
  2. Log in with your CERN credentials
  3. Spawn a server
    1. stay with all the default options and click "Start my session"
  4. Create a project first (which is basically creating a new directory) by clicking on the plus symbol on the top right
  5. When inside this directory, click the plus symbol again and choose Python 3
  6. Play around with it a bit, using python and ROOT syntax (don't forget to import ROOT)


FirstSteps.ipynb

  • No labels