...

In the following, the most important excerpts are described.

Annotated Description

...

#import libraries 
 
import uproot
import numpy as np
import pandas as pd
import h5py
import seaborn as sns
 
from sklearn.utils import shuffle
from sklearn.model_selection import train_test_split
from sklearn.datasets import make_classification
 
import tensorflow as tf
from tensorflow import keras
from tensorflow.keras.models import Sequential, Model
from tensorflow.keras.optimizers import SGD, Adam, RMSprop, Adagrad, Adadelta
from tensorflow.keras.layers import Input, Activation, Dense, Dropout
from tensorflow.keras.callbacks import EarlyStopping, ReduceLROnPlateau, ModelCheckpoint
from tensorflow.keras import utils
from tensorflow import random as tf_random
from keras.utils import plot_model
import random as python_random
 

[4]

# Fix random seed for reproducibility
 
# The below is necessary for starting Numpy generated random numbers
# in a well-defined initial state.
seed = 7
np.random.seed(seed)
 
# The below is necessary for starting core Python generated random numbers
# in a well-defined state.
python_random.seed(seed)
 
# The below set_seed() will make random number generation
# in the TensorFlow backend have a well-defined initial state.
# For further details, see: https://www.tensorflow.org/api_docs/python/tf/random/set_seed
tf_random.set_seed(seed)
 
treename = 'HZZ4LeptonsAnalysisReduced'
filename = {}
upfile = {}
params = {}
df = {}
 
# Define what are the ROOT files we are interested in (for the two categories,
# signal and background)
 
filename['sig'] = 'VBF_HToZZTo4mu.root'
filename['bkg_ggHtoZZto4mu'] = 'GluGluHToZZTo4mu.root'
filename['bkg_ZZto4mu'] = 'ZZTo4mu.root'
#filename['bkg_ttH_HToZZ_4mu.root']= 'ttH_HToZZ_4mu.root'
#filename['sig'] = 'VBF_HToZZTo4e.root'
#filename['bkg_ggHtoZZto4e'] = 'GluGluHToZZTo4e.root'
#filename['bkg_ZZto4e'] = 'ZZTo4e.root'
 
# Variables from Root Tree that must be copyed to PANDA dataframe (df)
VARS = [ 'f_run', 'f_event', 'f_weight', \
        'f_massjj', 'f_deltajj', 'f_mass4l', 'f_Z1mass' , 'f_Z2mass', \
        'f_lept1_pt','f_lept1_eta','f_lept1_phi', \
        'f_lept2_pt','f_lept2_eta','f_lept2_phi', \
        'f_lept3_pt','f_lept3_eta','f_lept3_phi', \
        'f_lept4_pt','f_lept4_eta','f_lept4_phi', \
        'f_jet1_pt','f_jet1_eta','f_jet1_phi', \
        'f_jet2_pt','f_jet2_eta','f_jet2_phi' ]
 
#checking the dimensions of the df , 26 variables
NDIM = len(VARS)
 
print("Number of kinematic variables imported from the ROOT files = %d"% NDIM)
 
upfile['sig'] = uproot.open(filename['sig'])
upfile['bkg_ggHtoZZto4mu'] = uproot.open(filename['bkg_ggHtoZZto4mu'])
upfile['bkg_ZZto4mu'] = uproot.open(filename['bkg_ZZto4mu'])
#upfile['bkg_ttH_HToZZ_4mu.root'] = uproot.open(filename['bkg_ttH_HToZZ_4mu'])
#upfile['sig'] = uproot.open(filename['sig'])]
#upfile['bkg_ggHtoZZto4e'] = uproot.open(filename['bkg_ggHtoZZto4e'])
#upfile['bkg_ZZto4e'] = uproot.open(filename['bkg_ZZto4e'])

Number of kinematic variables imported from the ROOT files = 26

Let's see what you have uploaded in your Colab notebook!

# Look at the signal and bkg events before applying physical requirement
 
df['sig'] = pd.DataFrame(upfile['sig'][treename].arrays(VARS, library="np"),columns=VARS)
print(df['sig'].shape)
 
 

(24867, 26)

Comment: We have 24867 rows, i.e. 24867 different events, and 26 columns (whose meaning will be explained later).

Let's print out the first rows of this data set!

df['sig'].head()

References

Attachments