Naive Bayes - S V M.txt

import csv
import numpy as np
import pandas as pd
mydata = pd.read_csv('IRIS.csv')

mydata.head(2)

mydata['species'].unique()

mydata.describe(include = 'all')

import seaborn as sns
import matplotlib.pyplot as plt
sns.pairplot(mydata)

g = sns.relplot( x = 'petal_length', y = 'petal_width', data = mydata, hue = 'species', style = 'species')
g.fig.set_size_inches(10, 5)
plt.show()

x = mydata.iloc[:, 0:4].values
y = mydata.iloc[:, 4].values

x

y

from sklearn.preprocessing import LabelEncoder
le = LabelEncoder()
y = le.fit_transform(y)

y

from sklearn.metrics import classification_report
from sklearn.metrics import confusion_matrix
from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score

from sklearn.model_selection import train_test_split
x_train, x_test, y_train, y_test = train_test_split(x, y, test_size = 0.3, random_state = 42)

Gaussian Naive Bayes

from sklearn.naive_bayes import GaussianNB
gaussian = GaussianNB()
gaussian.fit(x_train, y_train)
y_pred = gaussian.predict(x_test)
accuracy_nb = round(accuracy_score(y_test, y_pred)*100, 2)
acc_gaussian = round(gaussian.score(x_train, y_train)*100, 2)

cm = confusion_matrix(y_test, y_pred)
accuracy = accuracy_score(y_test, y_pred)
precision = precision_score(y_test, y_pred, average = 'micro')
recall = recall_score(y_test, y_pred, average = 'micro')
f1 = f1_score(y_test, y_pred, average = 'micro')

print('Confusion Matrix from Naive Bayes \n', cm)
print('Accuracy : %.3f' %accuracy)
print('Precision : %.3f' %precision)
print('Recall : %.3f' %recall)
print('F1score : %.3f' %f1)

Support Vector Machine

from sklearn.svm import SVC, LinearSVC

linear_svc = LinearSVC(max_iter = 4000)
linear_svc.fit(x_train, y_train)
y_pred = linear_svc.predict(x_test)
accuracy_svc = round(accuracy_score(y_test, y_pred)*100, 2)
acc_linear_svc = round(linear_svc.score(x_train, y_train)*100, 2)

cm = confusion_matrix(y_test, y_pred)
accuracy = accuracy_score(y_test, y_pred)
precision = precision_score(y_test, y_pred, average = 'micro')
recall = recall_score(y_test, y_pred, average = 'micro')
f1 = f1_score(y_test, y_pred, average = 'micro')

print('Confusion Matrix from Naive Bayes \n', cm)
print('Accuracy : %.3f' %accuracy)
print('Precision : %.3f' %precision)
print('Recall : %.3f' %recall)
print('F1score : %.3f' %f1)