Untitled

import math
import numpy as np
from scipy import stats
import matplotlib


#Q -> Total number of vehicles flowing per hour in the selected locations at different time intervals
#VP -> Percentage of heavy vehicles per hour in the selected locations at different time intervals

# Three sessions across 2 days each, a weekday and a weekend:
# morning (8:30am -8:40am)
# afternoon (1:00 pm -1:10 pm)
# evening (4:00 pm – 4:10pm)


#calculation of LAeq from the formula, given the values of Q and VP:
def CalixtoModelCalculation(Q, VP):
    Qeq = Q * (1 + 0.1 * VP)
    LAeq = 19.92224 * math.log10(Qeq) + 12.59764
    return LAeq

#function to calculate standard error:
def standard_error(data):
    return np.std(data, ddof=1) / np.sqrt(len(data))

#input arrays observed:
total_vehicles = [328, 358, 378, 224, 254, 252]
heavy_vehicles = [55, 68, 47, 36, 48, 51]

#arrays to store values:
Q_arr = []
VP_arr = []

ind=0
sz=len(total_vehicles)
while(ind < sz):
    Q_arr.append(total_vehicles[ind] * 6)
    VP_arr.append((heavy_vehicles[ind] / total_vehicles[ind]) * 100)
    ind = ind+1

#arrays to store Leq Values:
ActualLeq = []
PredictedLeq = []

#Noise Standards:
WHO = []
CPCB = []

#pointers required:
ind=0
sz = len(Q_arr)

while(ind < sz):
    #current values:
    Q = Q_arr[ind]
    VP = VP_arr[ind]

    #calculated values:
    Leq = CalixtoModelCalculation(Q, VP)
    PredictedLeq.append(Leq)

    #incremented pointer:
    ind = ind+1

#metrics used to test results:
correlation_coefficient, p_value = stats.pearsonr(ActualLeq, PredictedLeq)
f_statistic, p_value = stats.f_oneway(ActualLeq, PredictedLeq)
se_expected = standard_error(ActualLeq)
se_observed = standard_error(PredictedLeq)