/hbb-coffea

Clean repo for hbb studies

Primary LanguageJupyter Notebook

Set up environment by following instructions at https://github.com/CoffeaTeam/lpcjobqueue/

Make sure you have a valid grid certificate

Run over all input files for a particular year by doing

python submit-dask.py $year 

Run in a screen session or keep the terminal window open.

Follow the progress of dask jobs by opening another terminal window to the same cmlspc node:

ssh -L 8787:localhost:8787 USERNAME@cmslpcXXX@fnal.gov

Then point your browser to http://localhost:8787/

Coffea output files are put into the outfiles directory. Create a pickle of the histogram with name templates by doing

python make-pkl.py $year

Make root histograms to feed into combine using the python scripts in vh-scripts.