526_f.py

# -*- coding: utf-8 -*-
"""526_F.ipynb

Automatically generated by Colaboratory.

Original file is located at
    https://colab.research.google.com/drive/1AOYYHXUvCodecH5Hl6ZYLTwtPLiP-0GK
"""

#Importing the libraries
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns

# import warnings
import warnings
warnings.filterwarnings("ignore")

# We will use some methods from the sklearn module
from sklearn import linear_model
from sklearn.linear_model import LinearRegression
from sklearn import metrics
from sklearn.metrics import r2_score
from sklearn.metrics import mean_squared_error, mean_absolute_error
from sklearn.model_selection import train_test_split, cross_val_score

# Reading the Dataset
df = pd.read_csv("526_data_pointsN.csv")
df = df.iloc[100:]
df = df.iloc[:-100]
df.head()

df.shape

#Setting the value for X and Y
X = df[['branch-misses', 'cache-misses', 'L1-dcache-load-misses', 'L1-icache-load-misses', 'LLC-load-misses', 'LLC-store-misses', 'branch-load-misses', 'dTLB-load-misses', 'dTLB-store-misses', 'iTLB-load-misses', 'l2_rqsts.code_rd_miss', 'l2_rqsts.demand_data_rd_miss', 'l2_rqsts.all_demand_miss']]
i = df['instructions'].values.reshape(-1,1)
y = df['CPI']
X = np.divide(X,i)
df_new = X.copy()
df_new['CPI'] = y

corr = df.corr()
plt.figure (figsize = (15,8))
sns.heatmap(corr, annot = True)

print(df.columns.values.tolist())

X.drop(['LLC-store-misses','branch-load-misses','l2_rqsts.code_rd_miss','LLC-load-misses','dTLB-load-misses','L1-icache-load-misses'],axis=1,inplace = True)
X.head()

#Fitting the Multiple Linear Regression model
mlr = LinearRegression()

#Splitting dataset
X_train,X_test, y_train, y_test = train_test_split(X, y, test_size = 0.3, random_state = 100)

mlr.fit(X_train,y_train)

#Intercept and Coefficient
print("Intercept: ", mlr.intercept_)
print("Coefficients:")
list(zip(X, mlr.coef_))

#Prediction of test set
y_pred_mlr= mlr.predict(X_test)
#Predicted values
print("Prediction for test set: {}".format(y_pred_mlr))

#Actual value and the predicted value
mlr_diff = pd.DataFrame({'Actual value': y_test, 'Predicted value': y_pred_mlr})
mlr_diff.head()

#Model Evaluation
from sklearn import metrics
meanAbErr = metrics.mean_absolute_error(y_test, y_pred_mlr)
meanSqErr = metrics.mean_squared_error(y_test, y_pred_mlr)
rootMeanSqErr = np.sqrt(metrics.mean_squared_error(y_test, y_pred_mlr))
r2 = mlr.score(X_test,y_test)
print('R squared: {:.2f}'.format(r2*100))
print('Mean Absolute Error:', meanAbErr)
print('Mean Square Error:', meanSqErr)
print('Root Mean Square Error:', rootMeanSqErr)

final = X.mean()
R = np.multiply(final,mlr.coef_)
print(R)
print(R.sum()+mlr.intercept_)

finaly = y.mean()
print(finaly)

n = df.shape[0]
p = X.shape[1]

#Adjusted r2
Adjr2 = 1-(1-r2)*(n-1)/(n-p-1)
print(Adjr2)

#Residuals (added abs for now)
residuals = (y_test - y_pred_mlr)
print(residuals)

#F statistic
fstat = (r2/(1-r2))*((n-p-1)/p)
print(fstat)

from scipy.stats import f
p_value = 1-f.cdf(fstat,p,n-p-1)
print(p_value)

sns.set(style="whitegrid")
fig, ax = plt.subplots(figsize =(5,5))
sns.residplot(x=y_pred_mlr,y=residuals,ax=ax, lowess=True, line_kws={"color": "red"})
ax.set(ylabel='Residuals',xlabel='Predicted values')

import pandas as pd
R = pd.concat([pd.Series([mlr.intercept_],index = ['Base CPI']), R])
print(R)

fig, ax = plt.subplots()
groups = ['']
plt.figure (figsize = (2,5))
# Stacked bar chart with loop
for i in range(len(R)):
  plt.bar(groups, R[i],label = R.index[i], bottom = np.sum(R[:i], axis = 0),width = 0.1)

plt.legend(bbox_to_anchor = (1.25, 0.6), loc='upper left')
plt.tight_layout()