import pandas as pd
import numpy as np
from scipy import optimize

df = pd.read_csv('SolverPython.csv')

## Here I am trying to reduce the total error rate with the equation, error  = sum ((actual_value - predicted_value)/actual_value), where predicted value is given by the equation of linear regression(y = intercept + betas*x)

def equation(x, df):
    totalError = sum(abs(df['target'] - (66.75 - 0.9*df['variable1']* x[0]
                                        - 14.02*df['v2'] * x[1]
                                        - 2.57* df['v3'] * x[2]
                                        + 0.82 * df['v4'] * x[3]))/ df['target'])


    return totalError

x0 = [1, 1, 1, 1]  ### initialize the values of unknowns, which u want to optimize.

res = optimize.minimize(fun=equation, x0=np.array(x0), args=(df), method='SLSQP', bounds=[(0,20),(0,20),(0,20),(0,20)])


print(res)
print(res.x) ## to print the optimized values of x