keyboard_arrow_down Decision Tree

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
%matplotlib inline

df = pd.read_csv('/content/car_evaluation.csv', header=None)

df.shape

(1728, 7)

df.head()

0 1 2 3 4 5 6

0 vhigh vhigh 2 2 small low unacc

1 vhigh vhigh 2 2 small med unacc

2 vhigh vhigh 2 2 small high unacc

3 vhigh vhigh 2 2 med low unacc

4 vhigh vhigh 2 2 med med unacc

col_names = ['buying', 'maint', 'doors', 'persons', 'lug_boot', 'safety', 'class']

df.columns = col_names
col_names

['buying', 'maint', 'doors', 'persons', 'lug_boot', 'safety', 'class']

df.head()

buying maint doors persons lug_boot safety class

0 vhigh vhigh 2 2 small low unacc

1 vhigh vhigh 2 2 small med unacc

2 vhigh vhigh 2 2 small high unacc

3 vhigh vhigh 2 2 med low unacc

4 vhigh vhigh 2 2 med med unacc

 

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 1728 entries, 0 to 1727
Data columns (total 7 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 buying 1728 non-null object
1 maint 1728 non-null object
2 doors 1728 non-null object
3 persons 1728 non-null object
4 lug_boot 1728 non-null object
5 safety 1728 non-null object
6 class 1728 non-null object
dtypes: object(7)
memory usage: 94.6+ KB

Q. What is range index? Q. What does df.info() does?

col_names = ['buying', 'maint', 'doors', 'persons', 'lug_boot', 'safety', 'class']

for col in col_names:
print(df[col].value_counts())

buying
vhigh 432
high 432
med 432
low 432
Name: count, dtype: int64
 maint
vhigh 432
high 432
med 432
low 432
Name: count, dtype: int64
doors
2 432
3 432
4 432
5more 432
Name: count, dtype: int64
persons
2 576
4 576
more 576
Name: count, dtype: int64
lug_boot
small 576
med 576
big 576
Name: count, dtype: int64
safety
low 576
med 576
high 576
Name: count, dtype: int64
class
unacc 1210
acc 384
good 69
vgood 65
Name: count, dtype: int64

df['class'].value_counts()

count

class

unacc 1210

acc 384

good 69

vgood 65

dtype: int64
 

df.isnull().sum()

buying 0

maint 0

doors 0

persons 0

lug_boot 0

safety 0

class 0

dtype: int64
 

X = df.drop(['class'], axis=1)
y = df['class']

from sklearn.model_selection import train_test_split

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, random_state=42)

X_train.shape, X_test.shape

((1157, 6), (571, 6))

X_train.dtypes
 0

buying object

maint object

doors object

persons object

lug_boot object

safety object

dtype: object
 

!pip install category_encoders

Collecting category_encoders
Downloading category_encoders-2.6.3-py2.py3-none-any.whl.metadata (8.0 kB)
Requirement already satisfied: numpy>=1.14.0 in /usr/local/lib/python3.10/dist-packages (from category_encoders) (1.26.4)
Requirement already satisfied: scikit-learn>=0.20.0 in /usr/local/lib/python3.10/dist-packages (from category_encoders) (1.3.2)
Requirement already satisfied: scipy>=1.0.0 in /usr/local/lib/python3.10/dist-packages (from category_encoders) (1.13.1)
Requirement already satisfied: statsmodels>=0.9.0 in /usr/local/lib/python3.10/dist-packages (from category_encoders) (0.14.3)
Requirement already satisfied: pandas>=1.0.5 in /usr/local/lib/python3.10/dist-packages (from category_encoders) (2.1.4)
Requirement already satisfied: patsy>=0.5.1 in /usr/local/lib/python3.10/dist-packages (from category_encoders) (0.5.6)
Requirement already satisfied: python-dateutil>=2.8.2 in /usr/local/lib/python3.10/dist-packages (from pandas>=1.0.5->category_encod
Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas>=1.0.5->category_encoders) (2024
Requirement already satisfied: tzdata>=2022.1 in /usr/local/lib/python3.10/dist-packages (from pandas>=1.0.5->category_encoders) (20
Requirement already satisfied: six in /usr/local/lib/python3.10/dist-packages (from patsy>=0.5.1->category_encoders) (1.16.0)
Requirement already satisfied: joblib>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from scikit-learn>=0.20.0->category_encoder
Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from scikit-learn>=0.20.0->category_
Requirement already satisfied: packaging>=21.3 in /usr/local/lib/python3.10/dist-packages (from statsmodels>=0.9.0->category_encoder
Downloading category_encoders-2.6.3-py2.py3-none-any.whl (81 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 81.9/81.9 kB 4.9 MB/s eta 0:00:00
Installing collected packages: category_encoders
Successfully installed category_encoders-2.6.3

 

import category_encoders as ce

encoder = ce.OrdinalEncoder(cols=['buying', 'maint', 'doors', 'persons', 'lug_boot', 'safety'])

X_train = encoder.fit_transform(X_train)
X_test = encoder.transform(X_test)

X_train.head()

buying maint doors persons lug_boot safety

48 1 1 1 1 1 1

468 2 1 1 2 2 1

155 1 2 1 1 2 2

1721 3 3 2 1 2 2

1208 4 3 3 1 2 2

X_test.head()

buying maint doors persons lug_boot safety

599 2 2 4 3 1 2

1201 4 3 3 2 1 3

628 2 2 2 3 3 3

1498 3 2 2 2 1 3

1263 4 3 4 1 1 1

Using Gini Index

from sklearn.tree import DecisionTreeClassifier

clf_gini = DecisionTreeClassifier(criterion='gini', max_depth=3, random_state=0)
clf_gini.fit(X_train, y_train)

▾ DecisionTreeClassifier
DecisionTreeClassifier(max_depth=3, random_state=0)
y_pred_gini = clf_gini.predict(X_test)

from sklearn.metrics import accuracy_score

print('Model accuracy score with criterion gini index: {0:0.4f}'. format(accuracy_score(y_test, y_pred_gini)))

Model accuracy score with criterion gini index: 0.8021

y_pred_train_gini = clf_gini.predict(X_train)
y_pred_train_gini

array(['unacc', 'unacc', 'unacc', ..., 'unacc', 'unacc', 'acc'],

dtype=object)

print('Training set score: {:.4f}'.format(clf_gini.score(X_train, y_train)))

print('Test set score: {:.4f}'.format(clf_gini.score(X_test, y_test)))

Training set score: 0.7865

Test set score: 0.8021

Visualize Decision-Tree

plt.figure(figsize=(12,8))
from sklearn import tree
tree.plot_tree(clf_gini.fit(X_train, y_train))

[Text(0.4, 0.875, 'x[5] <= 1.5\ngini = 0.455\nsamples = 1157\nvalue = [255, 49, 813, 40]'),
Text(0.2, 0.625, 'gini = 0.0\nsamples = 386\nvalue = [0, 0, 386, 0]'),
Text(0.6, 0.625, 'x[3] <= 2.5\ngini = 0.577\nsamples = 771\nvalue = [255, 49, 427, 40]'),
Text(0.4, 0.375, 'x[0] <= 2.5\ngini = 0.631\nsamples = 525\nvalue = [255, 49, 181, 40]'),
Text(0.2, 0.125, 'gini = 0.496\nsamples = 271\nvalue = [124, 0, 147, 0]'),
Text(0.6, 0.125, 'gini = 0.654\nsamples = 254\nvalue = [131, 49, 34, 40]'),
Text(0.8, 0.375, 'gini = 0.0\nsamples = 246\nvalue = [0, 0, 246, 0]')]

import graphviz
dot_data = tree.export_graphviz(clf_gini, out_file=None,
feature_names=X_train.columns,
class_names=y_train,
filled=True)

graph = graphviz.Source(dot_data, format="png")

graph
 safety <= 1.5
gini = 0.455
samples = 1157
value = [255, 49, 813, 40]
class = unacc
False
True

persons <= 2.5

gini = 0.0
gini = 0.577
samples = 386
samples = 771
value = [0, 0, 386, 0]
value = [255, 49, 427, 40]
class = unacc
class = unacc

buying <= 2.5

gini = 0.0
gini = 0.631
samples = 246
samples = 525
value = [0, 0, 246, 0]
value = [255, 49, 181, 40]
class = unacc
class = unacc

gini = 0.496 gini = 0.654

samples = 271 samples = 254
value = [124, 0, 147, 0] value = [131, 49, 34, 40]
class = unacc class = unacc

Using Entropy

from sklearn.tree import DecisionTreeClassifier

clf_entropy = DecisionTreeClassifier(criterion='entropy', max_depth=3, random_state=0)
clf_entropy.fit(X_train, y_train)

▾ DecisionTreeClassifier
DecisionTreeClassifier(criterion='entropy', max_depth=3, random_state=0)

y_pred_entropy = clf_entropy.predict(X_test)
print('Model accuracy score with criterion entropy: {0:0.4f}'. format(accuracy_score(y_test, y_pred_entropy)))

Model accuracy score with criterion entropy: 0.8021

y_pred_train_entropy = clf_entropy.predict(X_train)
y_pred_train_entropy

array(['unacc', 'unacc', 'unacc', ..., 'unacc', 'unacc', 'acc'],

dtype=object)

accuracy_score(y_train, y_pred_train_entropy)

0.7865168539325843

print(clf_entropy.score(X_train, y_train))
print(clf_entropy.score(X_test, y_test))

0.7865168539325843
0.8021015761821366

plt.figure(figsize=(12,8))
from sklearn import tree
tree.plot_tree(clf_entropy.fit(X_train, y_train))
 [Text(0.4, 0.875, 'x[5] <= 1.5\nentropy = 1.2\nsamples = 1157\nvalue = [255, 49, 813, 40]'),
Text(0.2, 0.625, 'entropy = 0.0\nsamples = 386\nvalue = [0, 0, 386, 0]'),
Text(0.6, 0.625, 'x[3] <= 2.5\nentropy = 1.474\nsamples = 771\nvalue = [255, 49, 427, 40]'),
Text(0.4, 0.375, 'x[0] <= 2.5\nentropy = 1.638\nsamples = 525\nvalue = [255, 49, 181, 40]'),
Text(0.2, 0.125, 'entropy = 0.995\nsamples = 271\nvalue = [124, 0, 147, 0]'),
Text(0.6, 0.125, 'entropy = 1.759\nsamples = 254\nvalue = [131, 49, 34, 40]'),
Text(0.8, 0.375, 'entropy = 0.0\nsamples = 246\nvalue = [0, 0, 246, 0]')]

import graphviz
dot_data = tree.export_graphviz(clf_entropy, out_file=None,feature_names=X_train.columns,
class_names=y_train,filled=True)
graph

safety <= 1.5

gini = 0.455
samples = 1157
value = [255, 49, 813, 40]
class = unacc
False
True

persons <= 2.5

gini = 0.0
gini = 0.577
samples = 386
samples = 771
value = [0, 0, 386, 0]
value = [255, 49, 427, 40]
class = unacc
class = unacc

buying <= 2.5

gini = 0.0
gini = 0.631
samples = 246
samples = 525
value = [0, 0, 246, 0]
value = [255, 49, 181, 40]
class = unacc
class = unacc

gini = 0.496 gini = 0.654

samples = 271 samples = 254
value = [124, 0, 147, 0] value = [131, 49, 34, 40]
class = unacc class = unacc

Confusion Matrix
from sklearn.metrics import confusion_matrix

cm = confusion_matrix(y_test, y_pred_entropy)
print('Confusion matrix\n\n',cm)

Confusion matrix

[[ 73 0 56 0]
[ 20 0 0 0]
[ 12 0 385 0]
[ 25 0 0 0]]

Classification Report

from sklearn.metrics import classification_report

print(classification_report(y_test, y_pred_entropy))

precision recall f1-score support

acc 0.56 0.57 0.56 129

good 0.00 0.00 0.00 20
unacc 0.87 0.97 0.92 397
vgood 0.00 0.00 0.00 25

accuracy 0.80 571

macro avg 0.36 0.38 0.37 571
weighted avg 0.73 0.80 0.77 571

/usr/local/lib/python3.10/dist-packages/sklearn/metrics/_classification.py:1471: UndefinedMetricWarning: Precision and F-score are i

_warn_prf(average, modifier, msg_start, len(result))
/usr/local/lib/python3.10/dist-packages/sklearn/metrics/_classification.py:1471: UndefinedMetricWarning: Precision and F-score are i
_warn_prf(average, modifier, msg_start, len(result))
/usr/local/lib/python3.10/dist-packages/sklearn/metrics/_classification.py:1471: UndefinedMetricWarning: Precision and F-score are i
_warn_prf(average, modifier, msg_start, len(result))

 
keyboard_arrow_down Random Forest
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
%matplotlib inline

df = pd.read_csv('/content/car_evaluation.csv', header=None)

df.shape

(1728, 7)

df.head()

0 1 2 3 4 5 6

0 vhigh vhigh 2 2 small low unacc

1 vhigh vhigh 2 2 small med unacc

2 vhigh vhigh 2 2 small high unacc

3 vhigh vhigh 2 2 med low unacc

4 vhigh vhigh 2 2 med med unacc

col_names = ['buying', 'maint', 'doors', 'persons', 'lug_boot', 'safety', 'class']

df.columns = col_names
col_names

['buying', 'maint', 'doors', 'persons', 'lug_boot', 'safety', 'class']

df.head()

buying maint doors persons lug_boot safety class

0 vhigh vhigh 2 2 small low unacc

1 vhigh vhigh 2 2 small med unacc

2 vhigh vhigh 2 2 small high unacc

3 vhigh vhigh 2 2 med low unacc

4 vhigh vhigh 2 2 med med unacc

 

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 1728 entries, 0 to 1727
Data columns (total 7 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 buying 1728 non-null object
1 maint 1728 non-null object
2 doors 1728 non-null object
3 persons 1728 non-null object
4 lug_boot 1728 non-null object
5 safety 1728 non-null object
6 class 1728 non-null object
dtypes: object(7)
memory usage: 94.6+ KB

Q. What is range index? Q. What does df.info() does?

col_names = ['buying', 'maint', 'doors', 'persons', 'lug_boot', 'safety', 'class']

for col in col_names:
print(df[col].value_counts())

buying
vhigh 432
high 432
med 432
low 432
Name: count, dtype: int64
 maint
vhigh 432
high 432
med 432
low 432
Name: count, dtype: int64
doors
2 432
3 432
4 432
5more 432
Name: count, dtype: int64
persons
2 576
4 576
more 576
Name: count, dtype: int64
lug_boot
small 576
med 576
big 576
Name: count, dtype: int64
safety
low 576
med 576
high 576
Name: count, dtype: int64
class
unacc 1210
acc 384
good 69
vgood 65
Name: count, dtype: int64

df['class'].value_counts()

count

class

unacc 1210

acc 384

good 69

vgood 65

dtype: int64
 

df.isnull().sum()

buying 0

maint 0

doors 0

persons 0

lug_boot 0

safety 0

class 0

dtype: int64
 

X = df.drop(['class'], axis=1)
y = df['class']

from sklearn.model_selection import train_test_split

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, random_state=42)

X_train.shape, X_test.shape

((1157, 6), (571, 6))

X_train.dtypes
 0

buying object

maint object

doors object

persons object

lug_boot object

safety object

dtype: object
 

!pip install category_encoders

Collecting category_encoders
Downloading category_encoders-2.6.3-py2.py3-none-any.whl.metadata (8.0 kB)
Requirement already satisfied: numpy>=1.14.0 in /usr/local/lib/python3.10/dist-packages (from category_encoders) (1.26.4)
Requirement already satisfied: scikit-learn>=0.20.0 in /usr/local/lib/python3.10/dist-packages (from category_encoders) (1.3.2)
Requirement already satisfied: scipy>=1.0.0 in /usr/local/lib/python3.10/dist-packages (from category_encoders) (1.13.1)
Requirement already satisfied: statsmodels>=0.9.0 in /usr/local/lib/python3.10/dist-packages (from category_encoders) (0.14.3)
Requirement already satisfied: pandas>=1.0.5 in /usr/local/lib/python3.10/dist-packages (from category_encoders) (2.1.4)
Requirement already satisfied: patsy>=0.5.1 in /usr/local/lib/python3.10/dist-packages (from category_encoders) (0.5.6)
Requirement already satisfied: python-dateutil>=2.8.2 in /usr/local/lib/python3.10/dist-packages (from pandas>=1.0.5->category_encod
Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas>=1.0.5->category_encoders) (2024
Requirement already satisfied: tzdata>=2022.1 in /usr/local/lib/python3.10/dist-packages (from pandas>=1.0.5->category_encoders) (20
Requirement already satisfied: six in /usr/local/lib/python3.10/dist-packages (from patsy>=0.5.1->category_encoders) (1.16.0)
Requirement already satisfied: joblib>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from scikit-learn>=0.20.0->category_encoder
Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from scikit-learn>=0.20.0->category_
Requirement already satisfied: packaging>=21.3 in /usr/local/lib/python3.10/dist-packages (from statsmodels>=0.9.0->category_encoder
Downloading category_encoders-2.6.3-py2.py3-none-any.whl (81 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 81.9/81.9 kB 3.3 MB/s eta 0:00:00
Installing collected packages: category_encoders
Successfully installed category_encoders-2.6.3

 

import category_encoders as ce

encoder = ce.OrdinalEncoder(cols=['buying', 'maint', 'doors', 'persons', 'lug_boot', 'safety'])

X_train = encoder.fit_transform(X_train)
X_test = encoder.transform(X_test)

X_train.head()

buying maint doors persons lug_boot safety

48 1 1 1 1 1 1

468 2 1 1 2 2 1

155 1 2 1 1 2 2

1721 3 3 2 1 2 2

1208 4 3 3 1 2 2

X_test.head()

buying maint doors persons lug_boot safety

599 2 2 4 3 1 2

1201 4 3 3 2 1 3

628 2 2 2 3 3 3

1498 3 2 2 2 1 3

1263 4 3 4 1 1 1

Random Forest Classifier Model

from sklearn.ensemble import RandomForestClassifier

rfc = RandomForestClassifier(n_estimators=10,random_state=54)
rfc.fit(X_train, y_train)

▾ RandomForestClassifier
RandomForestClassifier(n_estimators=10, random_state=54)
y_pred = rfc.predict(X_test)

from sklearn.metrics import accuracy_score

print('Model accuracy score with 10 decision Trees', accuracy_score(y_test, y_pred))

Model accuracy score with 10 decision Trees 0.9369527145359019

rfc_100 = RandomForestClassifier(n_estimators=100,random_state=54)
rfc_100.fit(X_train, y_train)

▾ RandomForestClassifier
RandomForestClassifier(random_state=54)

y_pred = rfc_100.predict(X_test)

from sklearn.metrics import accuracy_score

print('Model accuracy score with 100 decision Trees', accuracy_score(y_test, y_pred))

Model accuracy score with 100 decision Trees 0.9474605954465849

clf= RandomForestClassifier(n_estimators=100,random_state=0)
clf.fit(X_train, y_train)

▾ RandomForestClassifier
RandomForestClassifier(random_state=0)

feature_scores = pd.Series(clf.feature_importances_, index=X_train.columns).sort_values(ascending=False)

feature_scores

safety 0.295319

persons 0.233856

buying 0.151734

maint 0.146653

lug_boot 0.100048

doors 0.072389

dtype: float64

sns.barplot(x=feature_scores, y=feature_scores.index)
plt.xlabel('Feature Importance Score')
plt.ylabel('Features')
plt.title("Visualizing Important Features")
plt.show()
Dropping the least imp features

X = df.drop(['class', 'doors'], axis=1)

y = df['class']

Again Splitting Train-Test as the least important feature has been dropped

from sklearn.model_selection import train_test_split

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, random_state=42)

encoder = ce.OrdinalEncoder(cols=['buying', 'maint', 'persons', 'lug_boot', 'safety'])

X_train = encoder.fit_transform(X_train)
X_test = encoder.transform(X_test)

rfc_100 = RandomForestClassifier(n_estimators=100,random_state=54)
rfc_100.fit(X_train, y_train)

▾ RandomForestClassifier
RandomForestClassifier(random_state=54)

y_pred = rfc_100.predict(X_test)

from sklearn.metrics import accuracy_score

print('Model accuracy score with 100 decision Trees is', accuracy_score(y_test, y_pred)*100,'%')

Model accuracy score with 100 decision Trees is 92.29422066549913 %

keyboard_arrow_down Cross Validation using K-Fold Cross Validation
from sklearn import datasets
from sklearn.tree import DecisionTreeClassifier
from sklearn.model_selection import KFold, cross_val_score

X, y = datasets.load_iris(return_X_y=True)

clf = DecisionTreeClassifier(random_state=42)

k_folds = KFold(n_splits = 5)

scores = cross_val_score(clf, X, y, cv = k_folds)

print("Cross Validation Scores: ", scores)

Cross Validation Scores: [1. 1. 0.83333333 0.93333333 0.8 ]

print("Average CV Score: ", scores.mean())

Average CV Score: 0.9133333333333333

print("Number of CV Scores used in Average: ", len(scores))

Number of CV Scores used in Average: 5

keyboard_arrow_down Heirarchical Clustering
import numpy as np # linear algebra
import pandas as pd

dataset = pd.read_csv('/content/Mall_Customers.csv')
dataset.head()

CustomerID Gender Age Annual Income (k$) Spending Score (1-100)

0 1 Male 19 15 39

1 2 Male 21 15 81

2 3 Female 20 16 6

3 4 Female 23 16 77

4 5 Female 31 17 40
 

Next steps: Generate code with dataset

toggle_off View recommended plots New interactive sheet

dataset.shape

(200, 5)

dataset.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 200 entries, 0 to 199
Data columns (total 5 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 CustomerID 200 non-null int64
1 Gender 200 non-null object
2 Age 200 non-null int64
3 Annual Income (k$) 200 non-null int64
4 Spending Score (1-100) 200 non-null int64
dtypes: int64(4), object(1)
memory usage: 7.9+ KB

dataset.describe()

CustomerID Age Annual Income (k$) Spending Score (1-100)

count 200.000000 200.000000 200.000000 200.000000

mean 100.500000 38.850000 60.560000 50.200000

std 57.879185 13.969007 26.264721 25.823522

min 1.000000 18.000000 15.000000 1.000000

25% 50.750000 28.750000 41.500000 34.750000

50% 100.500000 36.000000 61.500000 50.000000

75% 150.250000 49.000000 78.000000 73.000000

max 200.000000 70.000000 137.000000 99.000000

 

X = dataset.iloc[:, 3:]
X.head()

Annual Income (k$) Spending Score (1-100)

0 15 39

1 15 81

2 16 6

3 16 77

4 17 40
 

Next steps: Generate code with X toggle_off View recommended plots New interactive sheet
import scipy.cluster.hierarchy as hc
import matplotlib.pyplot as plt
from pylab import rcParams
rcParams['figure.figsize'] = 15, 10

# Using Dendogram to find the optimal number of clusters

dendogram = hc.dendrogram(hc.linkage(X, method = 'ward'))
plt.title('Dendrogram')
plt.xlabel('Customers')
plt.ylabel('Euclidean Distances')
plt.show()

dendogram = hc.dendrogram(hc.linkage(X, method = 'ward'))

plt.title('Dendrogram')
plt.xlabel('Customers')
plt.ylabel('Euclidean Distances')
plt.axhline(200, c='r', linestyle='--')
plt.show()
# Fitting hierarchical clustering to the mall dataset
from sklearn.cluster import AgglomerativeClustering
# Removing the affinity parameter as it is not used with 'ward' linkage.
hc_Agg = AgglomerativeClustering(n_clusters = 5, linkage = 'ward')
y_hc = hc_Agg.fit_predict(X)

Start coding or generate with AI.

Start coding or generate with AI.

Start coding or generate with AI.

Start coding or generate with AI.

# Visualizing the clusters

plt.scatter(X.iloc[y_hc == 0, 0], X.iloc[y_hc == 0, 1], s = 100, c = 'red', label = 'Careful')
plt.scatter(X.iloc[y_hc == 1, 0], X.iloc[y_hc == 1, 1], s = 100, c = 'blue', label = 'Standard')
plt.scatter(X.iloc[y_hc == 2, 0], X.iloc[y_hc == 2, 1], s = 100, c = 'green', label = 'Target')
plt.scatter(X.iloc[y_hc == 3, 0], X.iloc[y_hc == 3, 1], s = 100, c = 'cyan', label = 'Careless')
plt.scatter(X.iloc[y_hc == 4, 0], X.iloc[y_hc == 4, 1], s = 100, c = 'magenta', label = 'Sensible')
plt.title('Clusters of customers')
plt.xlabel('Annual Income (k$)')
plt.ylabel('Spending Score (1-100)')
plt.legend()
plt.show()
keyboard_arrow_down K-Means Clustering
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from sklearn.cluster import KMeans

df = pd.read_csv('/content/Mall_Customers.csv')
print("Shape of the data= ", df.shape)
df.head()

Shape of the data= (200, 5)

CustomerID Gender Age Annual Income (k$) Spending Score (1-100)

0 1 Male 19 15 39

1 2 Male 21 15 81

2 3 Female 20 16 6

3 4 Female 23 16 77

4 5 Female 31 17 40

plt.figure(figsize=(10,6))
plt.scatter(df['Annual Income (k$)'],df['Spending Score (1-100)'])
plt.xlabel('Annual Income')
plt.ylabel('Spending Score')
plt.title('Unlabelled Mall Customer Data')

Text(0.5, 1.0, 'Unlabelled Mall Customer Data')

 

# Since we are going to use Annual Income and Spending Score columns only, lets create 2D array of these columns for
X = df.iloc[:, [3,4]].values
X[:5] # Show first 5 records only

array([[15, 39],
[15, 81],
[16, 6],
[16, 77],
[17, 40]])

for each value of i, a kmeans model will be created with i number of clusters
init = random tells us about the random initialization of cluster centres
inertia_ = Sum of squared distances of samples to their closest cluster
adds or appends the inertia_ value to the clustering_score
clustering_score = [] #creating an empty list to store inertia values
for i in range(1, 11):
kmeans = KMeans(n_clusters = i, init = 'random', random_state = 42)
kmeans.fit(X)
clustering_score.append(kmeans.inertia_)
plt.figure(figsize=(10,6))
plt.plot(range(1, 11), clustering_score)
plt.scatter(5,clustering_score[4], s = 200, c = 'red', marker='*')
plt.title('The Elbow Method')
plt.xlabel('No. of Clusters')
plt.ylabel('Clustering Score')
plt.show()

 

kmeans= KMeans(n_clusters = 5, random_state = 42)

# Compute k-means clustering
kmeans.fit(X)
# Compute cluster centers and predict cluster index for each sample.
pred = kmeans.predict(X)
pred

array([4, 2, 4, 2, 4, 2, 4, 2, 4, 2, 4, 2, 4, 2, 4, 2, 4, 2, 4, 2, 4, 2,
4, 2, 4, 2, 4, 2, 4, 2, 4, 2, 4, 2, 4, 2, 4, 2, 4, 2, 4, 2, 4, 0,
4, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 3, 1, 0, 1, 3, 1, 3, 1,
0, 1, 3, 1, 3, 1, 3, 1, 3, 1, 0, 1, 3, 1, 3, 1, 3, 1, 3, 1, 3, 1,
3, 1, 3, 1, 3, 1, 3, 1, 3, 1, 3, 1, 3, 1, 3, 1, 3, 1, 3, 1, 3, 1,
3, 1, 3, 1, 3, 1, 3, 1, 3, 1, 3, 1, 3, 1, 3, 1, 3, 1, 3, 1, 3, 1,
3, 1], dtype=int32)

df['Cluster'] = pd.DataFrame(pred, columns=['cluster'] )

print('Number of data points in each cluster= \n', df['Cluster'].value_counts())
df
 Number of data points in each cluster=
Cluster
0 81
1 39
3 35
4 23
2 22
Name: count, dtype: int64
CustomerID Gender Age Annual Income (k$) Spending Score (1-100) Cluster

0 1 Male 19 15 39 4

1 2 Male 21 15 81 2

2 3 Female 20 16 6 4
plt.figure(figsize=(10,6))
3 4 Female 23 16 77 2
plt.scatter(X[pred == 0, 0], X[pred == 0, 1], c = 'brown', label = 'Cluster 0')
plt.scatter(X[pred
4 ==
5 1, 0], X[pred
Female 31 == 1, 1], c = 'green',
17 label = 'Cluster 40
1') 4
plt.scatter(X[pred == 2, 0], X[pred == 2, 1], c = 'blue', label = 'Cluster 2')
...
plt.scatter(X[pred ... 3, 0],... X[pred
== ... == 3, ...
1], c = 'purple', label = 'Cluster...3') ...
plt.scatter(X[pred == 4, 0], X[pred == 4, 1], c = 'orange', label = 'Cluster 4')
195 196 Female 35 120 79 1
plt.scatter(kmeans.cluster_centers_[:,0], kmeans.cluster_centers_[:, 1],s = 300, c = 'red', label = 'Centroid', marker='*')
plt.xlabel('Annual
196 Income')
197 Female 45 126 28 3
plt.ylabel('Spending Score')
197
plt.legend() 198 Male 32 126 74 1
plt.title('Customer
198 199Clusters')
Male 32 137 18 3

Text(0.5,
199 1.0,
200'Customer
Male Clusters')
30 137 83 1

200 rows × 6 columns

 
keyboard_arrow_down SVM
Classifying whether the customer is eligible to take a credit card or not

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
%matplotlib inline

df = pd.read_csv('/content/UniversalBank.csv')

df.describe()

Personal Securities
ID Age Experience Income ZIP Code Family CCAvg Education Mortgage
Loan Account

count 5000.000000 5000.000000 5000.000000 5000.000000 5000.000000 5000.000000 5000.000000 5000.000000 5000.000000 5000.000000 5000.000000 5

mean 2500.500000 45.338400 20.104600 73.774200 93152.503000 2.396400 1.937938 1.881000 56.498800 0.096000 0.104400

std 1443.520003 11.463166 11.467954 46.033729 2121.852197 1.147663 1.747659 0.839869 101.713802 0.294621 0.305809

min 1.000000 23.000000 -3.000000 8.000000 9307.000000 1.000000 0.000000 1.000000 0.000000 0.000000 0.000000

25% 1250.750000 35.000000 10.000000 39.000000 91911.000000 1.000000 0.700000 1.000000 0.000000 0.000000 0.000000

50% 2500.500000 45.000000 20.000000 64.000000 93437.000000 2.000000 1.500000 2.000000 0.000000 0.000000 0.000000

75% 3750.250000 55.000000 30.000000 98.000000 94608.000000 3.000000 2.500000 3.000000 101.000000 0.000000 0.000000

max 5000.000000 67.000000 43.000000 224.000000 96651.000000 4.000000 10.000000 3.000000 635.000000 1.000000 1.000000

df.head()

ID Age Experience Income ZIP Code Family CCAvg Education Mortgage Personal Loan Securities Account CD Account Online CreditCard

0 1 25 1 49 91107 4 1.6 1 0 0 1 0 0 0

1 2 45 19 34 90089 3 1.5 1 0 0 1 0 0 0

2 3 39 15 11 94720 1 1.0 1 0 0 0 0 0 0

3 4 35 9 100 94112 1 2.7 2 0 0 0 0 0 0

4 5 35 8 45 91330 4 1.0 2 0 0 0 0 0 1
 

df.isnull().sum()

ID 0

Age 0

Experience 0

Income 0

ZIP Code 0

Family 0

CCAvg 0

Education 0

Mortgage 0

Personal Loan 0

Securities Account 0

CD Account 0

Online 0

CreditCard 0

dtype: int64
 

df1 = df.drop(['ID','ZIP Code'],axis=1)

df1.head()
 Age Experience Income Family CCAvg Education Mortgage Personal Loan Securities Account CD Account Online CreditCard

0 25 1 49 4 1.6 1 0 0 1 0 0 0

1 45 19 34 3 1.5 1 0 0 1 0 0 0

2 39 15 11 1 1.0 1 0 0 0 0 0 0

3 35 9 100 1 2.7 2 0 0 0 0 0 0

4 35 8 45 4 1.0 2 0 0 0 0 0 1
 

plt.figure(figsize=(10,5))
plt.title('Heatmap showing correlation between all the features', fontsize=10)
sns.heatmap(df1.corr(),annot=True, cmap= 'mako')

<Axes: title={'center': 'Heatmap showing correlation between all the features'}>

 

separating class 0 and class 1

zero_class = df1[df1['CreditCard']==0]
one_class = df1[df1['CreditCard']==1]

zero_class.head()

Age Experience Income Family CCAvg Education Mortgage Personal Loan Securities Account CD Account Online CreditCard

0 25 1 49 4 1.6 1 0 0 1 0 0 0

1 45 19 34 3 1.5 1 0 0 1 0 0 0

2 39 15 11 1 1.0 1 0 0 0 0 0 0

3 35 9 100 1 2.7 2 0 0 0 0 0 0

5 37 13 29 4 0.4 2 155 0 0 0 1 0
 

one_class.head()

Age Experience Income Family CCAvg Education Mortgage Personal Loan Securities Account CD Account Online CreditCard

4 35 8 45 4 1.0 2 0 0 0 0 0 1

7 50 24 22 1 0.3 3 0 0 0 0 0 1

15 60 30 22 1 1.5 3 0 0 0 0 1 1

19 55 28 21 1 0.5 2 0 0 1 0 0 1

24 36 11 152 2 3.9 1 159 0 0 0 0 1

 

Scatter Plot

plt.figure(figsize=(10,10)) # Set the figure size using the 'figure' function

plt.xlabel('Income')
plt.ylabel('Experience')
plt.scatter(zero_class['Income'],zero_class['Experience'],color='green',marker='+')
plt.scatter(one_class['Income'],one_class['Experience'],color='red',marker='.')
 <matplotlib.collections.PathCollection at 0x7edb6ace2740>

 

plt.xlabel('CCAvg')
plt.ylabel('Family')
plt.scatter(zero_class['CCAvg'],zero_class['Family'],color='blue',marker='+')
plt.scatter(one_class['CCAvg'],one_class['Family'],color='red',marker='.')

<matplotlib.collections.PathCollection at 0x7edb67282950>

 

Standard Scaler

from sklearn.preprocessing import StandardScaler

scaler = StandardScaler()
scaled = scaler.fit_transform(df1.drop('CreditCard',axis=1))
df_scaled = pd.DataFrame(scaled,columns=df1.columns[:-1])
df_scaled.head()
 Age Experience Income Family CCAvg Education Mortgage Personal Loan Securities Account CD Account Online

0 -1.774417 -1.666078 -0.538229 1.397414 -0.193385 -1.049078 -0.555524 -0.325875 2.928915 -0.25354 -1.216618

1 -0.029524 -0.096330 -0.864109 0.525991 -0.250611 -1.049078 -0.555524 -0.325875 2.928915 -0.25354 -1.216618

2 -0.552992 -0.445163 -1.363793 -1.216855 -0.536736 -1.049078 -0.555524 -0.325875 -0.341423 -0.25354 -1.216618

3 -0.901970 -0.968413 0.569765 -1.216855 0.436091 0.141703 -0.555524 -0.325875 -0.341423 -0.25354 -1.216618

4 -0.901970 -1.055621 -0.625130 1.397414 -0.536736 0.141703 -0.555524 -0.325875 -0.341423 -0.25354 -1.216618
 

x = df_scaled
y = df1['CreditCard']

from sklearn.model_selection import train_test_split

x_train,x_test,y_train,y_test = train_test_split(x,y,test_size=0.2,random_state=42)

#apply SVM model

from sklearn.svm import SVC
from sklearn.metrics import accuracy_score
model = SVC()
model.fit(x_train,y_train)

▾ SVC i ?
SVC()
 

y_pred=model.predict(x_test)
print('Model accuracy : {0:0.3f}%'.format(accuracy_score(y_test,y_pred)*100))

Model accuracy : 75.000%

from sklearn.metrics import confusion_matrix

cm = confusion_matrix(y_test,y_pred)
cm_matrix = pd.DataFrame(data=cm,columns=['Predicted:0','Predicted:1'],index=['Actual:0','Actual:1'])
sns.heatmap(cm_matrix,annot=True,fmt='d',cmap='mako')

<Axes: >

 

from sklearn.metrics import confusion_matrix

cm = confusion_matrix(y_test,y_pred)
cm

array([[703, 5],
[245, 47]])

#classification report
from sklearn.metrics import classification_report
print(classification_report(y_test,y_pred))

precision recall f1-score support

0 0.74 0.99 0.85 708

1 0.90 0.16 0.27 292

accuracy 0.75 1000

macro avg 0.82 0.58 0.56 1000
weighted avg 0.79 0.75 0.68 1000

Applying Linear Kernal

linear_classifier = SVC(kernel='linear').fit(x_train,y_train)
y_pred = linear_classifier.predict(x_test)
print('Model accuracy : {0:0.3f}%'.format(accuracy_score(y_test,y_pred)*100))

Model accuracy : 75.200%

cm = confusion_matrix(y_test,y_pred)
cm_matrix = pd.DataFrame(data=cm,columns=['Predicted:0','Predicted:1'],index=['Actual:0','Actual:1'])
sns.heatmap(cm_matrix,annot=True,fmt='d',cmap='mako')

<Axes: >

 

Gaussian RBF kernel

rbf_svc=SVC(kernel='rbf').fit(x_train,y_train)
y_pred=rbf_svc.predict(x_test)
print('Model accuracy : {0:0.3f}%'.format(accuracy_score(y_test,y_pred)*100))

Model accuracy : 75.000%

cm = confusion_matrix(y_test,y_pred)
cm_matrix = pd.DataFrame(data=cm,columns=['Predicted:0','Predicted:1'],index=['Actual:0','Actual:1'])
sns.heatmap(cm_matrix,annot=True,fmt='d',cmap='mako')

<Axes: >

 

Polynomial Kernel Function

Poly_svc = SVC(kernel='poly').fit(x_train,y_train)
y_pred = Poly_svc.predict(x_test)

print('Model accuracy : {0:0.3f}%'.format(accuracy_score(y_test,y_pred)*100))

Model accuracy : 75.000%

cm = confusion_matrix(y_test,y_pred)
cm_matrix = pd.DataFrame(data=cm,columns=['Predicted:0','Predicted:1'],index=['Actual:0','Actual:1'])
sns.heatmap(cm_matrix,annot=True,fmt='d',cmap='mako')
 <Axes: >

 

Sigmoid Kernal Function

Poly_svc = SVC(kernel='sigmoid').fit(x_train,y_train)
y_pred = Poly_svc.predict(x_test)
print('Model accuracy : {0:0.3f}%'.format(accuracy_score(y_test,y_pred)*100))

Model accuracy : 60.400%

cm = confusion_matrix(y_test,y_pred)
cm_matrix = pd.DataFrame(data=cm,columns=['Predicted:0','Predicted:1'],index=['Actual:0','Actual:1'])
sns.heatmap(cm_matrix,annot=True,fmt='d',cmap='mako')

<Axes: >

 