Coffea and bacon

A repository for using columnar analysis on bacon(bits) files.

Setup

To use, first an environment must be setup. If running on your laptop, running pip install coffea jupyter should get most of the way there. It might be preferable to install some of these things via conda. Note, this is a python 3 package; python 2 is end of life so get used to print(). To install at the LPC, there is a script:

source setup_lcg.sh

On future use at LPC, run source env_lcg.sh.

Running the Hbb analysis on baconbits

The following recipe runs all the relevant code to produce templates similar to those of sampleContainer:

cd analysis
# optional, because output saved in repository: ./make_pileup.py
./compile_corrections.py
./boostedHbbProcessor.py
./run_baconbits.py --executor futures --sample Hbb_2017
python baconbits-templates.py
python convert2d.py
ls hist_1DZbb*

This will take about 25 minutes to run. To just get your feet wet, look at ./run_baconbits.py --help, then run

./download_testbits.sh
./run_baconbits.py --sample test_bits

which will not take much time.

Plots with jupyter

To use jupyter on a laptop, jupyter notebook should work. To use juptyer at LPC, start a notebook server with, e.g. jupyter notebook --no-browser --port $PORT, substituing your favorite port: pick a random integer in (8000,65535). Often, you'll need an ssh tunnel, which can be accomplished via, e.g. ssh -L $PORT:localhost:$PORT server.address

Check out some of the notebooks in the analysis directory. Most will need the hists.coffea file as created by run_baconbits.py, but the top background notebook can be run after downloading the test baconbits.