demo1.py

Go to the documentation of this file.

import os
import sys
sys.path.append('../..')

from PandAna.core import *
import matplotlib.pyplot as plt

# Better var
def kPngE(tables):
    df = tables['rec.vtx.elastic.fuzzyk.png']['calE']
    return df.groupby(level = KL).sum()
kPngE = Var(kPngE)
# NOTE: No check that there are actually prongs in the event
# The rec.vtx.elastic.fuzzyk.png dataset already only has events with prongs

# Vertex fiducial cut
def kFiducial(tables):
    df = tables['rec.vtx.elastic']
    return (df['vtx.x'] < 180) & \
        (df['vtx.x'] > -180) & \
        (df['vtx.y'] < 180) & \
        (df['vtx.y'] > -180) & \
        (df['vtx.z'] < 1000) & \
        (df['vtx.z'] > 50)
kFiducial = Cut(kFiducial)

if __name__ == '__main__':
    # Latest h5s from Karl
    loc = '/pnfs/nova/persistent/users/karlwarb/HDF5-Training-19-02-26/FD-FluxSwap-FHC'
    files = [os.path.join(loc, f) for f in os.listdir(loc) if 'h5caf.h5' in f]
    tables = loader(files, limit=100)

    # Create a spectrum
    myspectrum = spectrum(tables, kFiducial, kPngE)

    # Let's do it!
    tables.Go()

    print('myspectrum internal dataframe: ')
    print(myspectrum.df().head())

    n, bins = myspectrum.histogram(bins=20, range=(1,4))

    print('Selected '+ str(n.sum()) + ' events')

    plt.hist(bins[:-1], bins=bins, weights=n)

    plt.xlabel('Prong Energy')
    plt.ylabel('Events')

    plt.legend(loc='upper right')

    plt.show()