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@[email protected]
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.