from sklearn.model_selection import train_test_split

from sklearn.linear_model import LogisticRegression

from sklearn.ensemble import RandomForestClassifier

from sklearn.svm import SVC

#importing metric that here is accuracy_score

from sklearn.metrics import accuracy_score

import pandas as pd

dataset=pd.read_csv("dataset.csv")

X = dataset.iloc[:,1:].values

y = dataset.iloc[:,0].values

i=0

for i in range(128):

a=list(y[i].split("_"))

dataset=dataset.set_value(i,'sample',a[1]+"_"+a[2])

#converting categorical data to Numerical

from sklearn.preprocessing import LabelEncoder

label_encoder = LabelEncoder()

dataset.iloc[:,0] = label_encoder.fit_transform(dataset.iloc[:,0].values)

y=dataset.iloc[:,0].values

#splitting data in 80:20 ratio

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 0)

#applying LogisticRegression

lr = LogisticRegression(C=10.0)

lr.fit(X_train, y_train)

#applying Random Forest Classsifier

rf = RandomForestClassifier(max_depth=None)

rf.fit(X_train, y_train)

#Applying SVM

svc = SVC(C=5.0)

svc.fit(X_train, y_train)

column=list(dataset.columns.values)

column.remove('sample')

value=[]

value1=[]

value2=[]

#Predicting values

a=lr.predict(X_test)

b=rf.predict(X_test)

c=svc.predict(X_test)

#Converting numerical to categorical for spacing what class does prediction belong to

for i in a:

if i==0:

value.append("babble_sn10")

elif i==1:

value.append("babble_sn5")

elif i==2:

value.append("car_sn10")

elif i==3:

value.append("car_sn5")

elif i==4:

value.append("street_sn10")

elif i==5:

value.append("street_sn5")

elif i==6:

value.append("train_sn10")

elif i==7:

value.append("train_sn5")

for i in b:

if i==0:

value1.append("babble_sn10")

elif i==1:

value1.append("babble_sn5")

elif i==2:

value1.append("car_sn10")

elif i==3:

value1.append("car_sn5")

elif i==4:

value1.append("street_sn10")

elif i==5:

value1.append("street_sn5")

elif i==6:

value1.append("train_sn10")

elif i==7:

value1.append("train_sn5")

for i in c:

if i==0:

value2.append("babble_sn10")

elif i==1:

value2.append("babble_sn5")

elif i==2:

value2.append("car_sn10")

elif i==3:

value2.append("car_sn5")

elif i==4:

value2.append("street_sn10")

elif i==5:

value2.append("street_sn5")

elif i==6:

value2.append("train_sn10")

elif i==7:

value2.append("train_sn5")

value3=[]

for i in y_test:

if i==0:

value3.append("babble_sn10")

elif i==1:

value3.append("babble_sn5")

elif i==2:

value3.append("car_sn10")

elif i==3:

value3.append("car_sn5")

elif i==4:

value3.append("street_sn10")

elif i==5:

value3.append("street_sn5")

elif i==6:

value3.append("train_sn10")

elif i==7:

value3.append("train_sn5")

#Comparing predicted vs actual

print()

print("-----------------------------------------")

print("| Logistic Regression Predictions |")

print("-----------------------------------------")

for i in range(26):

print()

print("Predicted: ",value[i]," ","Actual: ",value3[i])

print()

print("-----------------------------------------")

print("|Random Forest Classification Predictions|")

print("-----------------------------------------")

for i in range(26):

print()

print("Predicted: ",value1[i]," ","Actual: ",value3[i])

print()

print("-----------------------------------------")

print("| SVM Predictions |")

print("-----------------------------------------")

for i in range(26):

print()

print("Predicted: ",value2[i]," ","Actual: ",value3[i])

print("-----------------------------------------")

print("| Final Result |")

print("-----------------------------------------")

print()

print("Logistic Regression: ", accuracy_score(y_test, a)*100)

print()

print("Random Forest Classifier: ", accuracy_score(y_test,b)*100)

print()

print("Support Vector Machine: ", accuracy_score(y_test,c)*100)

print()