validation.py

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
 
 
 
"""\
This module contains classes for plotting validation results.
"""
import pxd.utils as utils
from pxd.utils.plots import *
import uproot
import ROOT
import pandas as pd
import numpy as np
 
 
def analyse_root_file(file_name, tree_name="tree"):
    """
    Function to analyse the ROOT file generated from validation algorithm
    Parameters:
      file_name (str): name of the ROOT file
      tree_name (str): name of the TTree
    """
    # Get TTree
    tfile = ROOT.TFile.Open(file_name)
    tree = tfile.Get(tree_name)
    # Read pandas.DataFrame from some branches
    branches = ['exp', 'run', 'pxdid', 'uBin', 'vBin', 'nTrackPoints', 'nSelTrackPoints', 'nSelTrackClusters', 'nTrackClusters']
    df = uproot.open(file_name)[tree_name].arrays(branches, library='pd')
    experiment_number = df.exp.min()
    run_range = (df.run.min(), df.run.max())
    prefix = f"e{experiment_number:05}_r{run_range[0]:04}-{run_range[1]:04}_"
 
    # Calculate efficiencies for all regions
    df.calculate_eff(num="nTrackClusters", den="nTrackPoints", output_var="eff")
    df.calculate_eff(num="nSelTrackClusters", den="nSelTrackPoints", output_var="eff_sel")
 
    plot_in_module_efficiency(
        df=df,
        pxdid=1052,
        y="eff_sel",
        x="run",
        cut="eff_sel>0&eff_sel_err_low<0.01",
        yerr_low="eff_sel_err_low",
        yerr_up="eff_sel_err_up",
        save_to=prefix + "efficiency_in_selected_regions_H1052.png")
 
    # Calculate efficiency of the entire PXD
    df_sum_all = df.groupby(['exp', 'run'])[["nTrackPoints", "nSelTrackPoints",
                                             "nTrackClusters", "nSelTrackClusters"]].sum().reset_index()
    df_sum_all.calculate_eff(num="nTrackClusters", den="nTrackPoints", output_var="eff")
    df_sum_all.calculate_eff(num="nSelTrackClusters", den="nSelTrackPoints", output_var="eff_sel")
 
    plot_efficiency_vs_run(df=df_sum_all, max_err=0.01, save_to=prefix + "pxd_efficiency_vs_run.png")
 
    # Calculate efficiences of modules
    df_module_sum = df.groupby(['exp', 'run', 'pxdid'])[["nTrackPoints", "nSelTrackPoints",
                                                         "nTrackClusters", "nSelTrackClusters"]].sum().reset_index()
    df_module_sum["pxdid"] = df_module_sum["pxdid"].astype(np.int)
    df_module_sum.calculate_eff(num="nTrackClusters", den="nTrackPoints", output_var="eff")
    df_module_sum.calculate_eff(num="nSelTrackClusters", den="nSelTrackPoints", output_var="eff_sel")
 
    plot_module_efficiencies_in_DHHs(df=df_module_sum, save_to=prefix + "efficiency_vs_run.png")
 
    # 2d efficiency map
    c1 = ROOT.TCanvas("c1", "c1", 900, 600)
    ROOT.gStyle.SetPalette(ROOT.kDarkRainBow)
    h2eff_l1 = plot_efficiency_map(
        tree=tree,
        exp_runs=[
            0,
            9999],
        num_name="hTrackClustersLayer1",
        den_name="hTrackPointsLayer1",
        save_to=prefix + "efficiency_l1.png",
        canvas=c1)
    h2eff_l2 = plot_efficiency_map(
        tree=tree,
        exp_runs=[
            0,
            9999],
        num_name="hTrackClustersLayer2",
        den_name="hTrackPointsLayer2",
        title="Efficiency of layer 2",
        save_to=prefix + "efficiency_l2.png",
        canvas=c1)
    # ip resolutions
    df_res = plot_ip_resolutions(tree=tree, save_to=prefix + "pxd_resolutions_vs_run.png")
 
 
if __name__ == "__main__":
    import sys
    file_name = sys.argv[1]
    ROOT.gROOT.SetBatch(True)
    analyse_root_file(file_name)