ML4DQM-Visualizations

This repo contains the code necessary to produces all the histograms for each quantile and each run for given datasets. Just copy the allplots.py file in three separate folder day good, bad, bad_dcs then for each copy of this code change the directory of the dataset for good, bad , bad_dcs use command

1. nohup python allplots.py > good.log & # inside the good folder
2. nohup python allplots.py > bad.log & # inside the bad folder
3. nohup python allplots.py > bad.log & # inside the bad_dcs folder

Jupyter notebook access on the minsky cluster

0. I used chrome for this to work
1. ssh -C -D 9999 [email protected]
2. change the proxy settings

3.http://ibmminsky-head.cern.ch:8080/platform

4.login with your cern credentials

5. Workload> Spark > My apps and Notebooks [ this NB usually doesn't work when I wrote these insturctions]

6. So,go to http://ibmminsky-3.cern.ch:9970

7. Jupyter NB Tutorial - https://www.ibm.com/support/knowledgecenter/en/SSZU2E_2.3.0/tutorial/t_notebook.html

8. ref this pdf from the minsky knowledge tranfer event https://github.com/rishabhCMS/ML4DQM-Visualizations/blob/master/DataViz2018/Presentations/20190903-MinskyClusterTraining.pdf

9. once you are inside the notebook you can chnage the directory using !cd and !pwd etc