GenSet_2.py

# -*- coding: utf-8 -*-
"""
Created on Mon Aug 10 15:36:29 2020

@author: Dell
"""
import pandas as pd
from sklearn.utils import shuffle
from sklearn import linear_model
from sklearn.model_selection import cross_val_score, cross_validate, cross_val_predict
import numpy as np
from sklearn.gaussian_process.kernels import RBF
from sklearn.gaussian_process import GaussianProcessRegressor
from math import sqrt as sq
import matplotlib.pyplot as plt
import time
from sklearn.model_selection import train_test_split
from sklearn.metrics import mean_absolute_error, mean_squared_error, r2_score
from joblib import dump,load
#%%

data = pd.read_excel('Data_Base.xls', index_col=0, Header=None)   


y = pd.DataFrame()
target= 'Max Demand' #  'Renewable Capacity' 'Renewable Penetration'
y[target] = data[target]*0.75

y=y.astype('float')

X = pd.DataFrame()
X['Renewable Invesment Cost'] = data['Renewable Unitary Invesment Cost']   
X['Battery Unitary Invesment Cost'] = data['Battery Unitary Invesment Cost']
X['Deep of Discharge'] = data['Deep of Discharge']
X['Battery Cycles'] = data['Battery Cycles']
X['GenSet Unitary Invesment Cost'] = data['GenSet Unitary Invesment Cost']
X['Generator Efficiency'] = data['Generator Efficiency']
X['Low Heating Value'] = data['Low Heating Value']
X['Fuel Cost'] = data['Fuel Cost']
#X['Generator Nominal capacity'] = data['Generator Nominal capacity'] 
X['HouseHolds'] = data['HouseHolds']
X['Renewable Energy Unit Total'] = data['Renewable Energy Unit Total']
#X['Max Demand'] = data['Max Demand']


#%%

scoring =   ['r2', 'neg_mean_absolute_error', 'neg_mean_squared_error'] #'r2' 'neg_mean_absolute_error' # 'neg_mean_squared_error'
for i in scoring:
    
    lm = linear_model.LinearRegression(fit_intercept=False)
    scores = cross_val_score(lm, X, y, cv=5,scoring=i)
    score = round(scores.mean(),2)
    
    if i == 'neg_mean_squared_error':
        score = sq(-score)    
        print(i + ' for the gaussian process with the test data set is ' + str(score))
    else:    
        print(i + ' for the gaussian process with the test data set is ' + str(score))
    
#%%
    


lm1= linear_model.LinearRegression(fit_intercept=True)
lm1 = lm1.fit(X,y)
dump(lm1, 'GenSet_Chaco.joblib')