10/1/24, 11:26 PM Ensmble_Learning_ML_5 - Jupyter Notebook

In [1]: #bagging - bootstrap aggregating, parallel approach

#boosting - sequential approach
#decision tree - supervised learning, non-parametric
#To avoid the drawbrack of decision tree i.e ovefitting, random forest mode
#Random forest is used for feature selection

In [1]: import pandas as pd

import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score, confusion_matrix, classificatio
from sklearn.tree import DecisionTreeClassifier
from sklearn.ensemble import RandomForestClassifier
from sklearn.tree import plot_tree

In [2]: df = pd.read_csv('car_data.csv')

In [3]: df.head()

Out[3]:
vhigh vhigh.1 2 2.1 small low unacc

0 vhigh vhigh 2 2 small med unacc

1 vhigh vhigh 2 2 small high unacc

2 vhigh vhigh 2 2 med low unacc

3 vhigh vhigh 2 2 med med unacc

4 vhigh vhigh 2 2 med high unacc

In [4]: df.shape

Out[4]: (1727, 7)

In [5]: df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 1727 entries, 0 to 1726
Data columns (total 7 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 vhigh 1727 non-null object
1 vhigh.1 1727 non-null object
2 2 1727 non-null object
3 2.1 1727 non-null object
4 small 1727 non-null object
5 low 1727 non-null object
6 unacc 1727 non-null object
dtypes: object(7)
memory usage: 94.6+ KB

localhost:8888/notebooks/BE_PRACTICALS/Ensmble_Learning_ML_5.ipynb 1/7
10/1/24, 11:26 PM Ensmble_Learning_ML_5 - Jupyter Notebook

In [6]: df.describe()

Out[6]:
vhigh vhigh.1 2 2.1 small low unacc

count 1727 1727 1727 1727 1727 1727 1727

unique 4 4 4 3 3 3 4

top high high 3 4 med med unacc

freq 432 432 432 576 576 576 1209

In [7]: df.isna().sum()

Out[7]: vhigh 0
vhigh.1 0
2 0
2.1 0
small 0
low 0
unacc 0
dtype: int64

In [8]: df_col = ['buying', 'maint', 'doors', 'persons', 'lug_boot', 'safety', 'cla

df.columns=df_col
df.head()

Out[8]:
buying maint doors persons lug_boot safety class

0 vhigh vhigh 2 2 small med unacc

1 vhigh vhigh 2 2 small high unacc

2 vhigh vhigh 2 2 med low unacc

3 vhigh vhigh 2 2 med med unacc

4 vhigh vhigh 2 2 med high unacc

localhost:8888/notebooks/BE_PRACTICALS/Ensmble_Learning_ML_5.ipynb 2/7
10/1/24, 11:26 PM Ensmble_Learning_ML_5 - Jupyter Notebook

In [9]: for i in df_col:

print(df[i].value_counts())

buying
high 432
med 432
low 432
vhigh 431
Name: count, dtype: int64
maint
high 432
med 432
low 432
vhigh 431
Name: count, dtype: int64
doors
3 432
4 432
5more 432
2 431
Name: count, dtype: int64
persons
4 576
more 576
2 575
Name: count, dtype: int64
lug_boot
med 576
big 576
small 575
Name: count, dtype: int64
safety
med 576
high 576
low 575
Name: count, dtype: int64
class
unacc 1209
acc 384
good 69
vgood 65
Name: count, dtype: int64

In [10]: df['class'].unique()

Out[10]: array(['unacc', 'acc', 'vgood', 'good'], dtype=object)

In [11]: x = df.drop('class', axis = 1)

y = df['class']
y.unique()

Out[11]: array(['unacc', 'acc', 'vgood', 'good'], dtype=object)

In [12]: x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.2, ra

localhost:8888/notebooks/BE_PRACTICALS/Ensmble_Learning_ML_5.ipynb 3/7
10/1/24, 11:26 PM Ensmble_Learning_ML_5 - Jupyter Notebook

In [13]: x_train.head()

Out[13]:
buying maint doors persons lug_boot safety

107 vhigh high 2 2 small low

900 med vhigh 3 4 small med

1708 low low 5more 2 big high

705 high med 4 2 med med

678 high med 3 2 med med

In [14]: x_train.shape

Out[14]: (1381, 6)

In [15]: x_test.shape

Out[15]: (346, 6)

In [16]: x_train.dtypes

Out[16]: buying object

maint object
doors object
persons object
lug_boot object
safety object
dtype: object

localhost:8888/notebooks/BE_PRACTICALS/Ensmble_Learning_ML_5.ipynb 4/7
10/1/24, 11:26 PM Ensmble_Learning_ML_5 - Jupyter Notebook

In [17]: !pip install category_encoders

Collecting category_encoders
Downloading category_encoders-2.6.3-py2.py3-none-any.whl.metadata (8.0 k
B)
Requirement already satisfied: numpy>=1.14.0 in c:\python310\lib\site-pack
ages (from category_encoders) (1.23.5)
Requirement already satisfied: scikit-learn>=0.20.0 in c:\python310\lib\si
te-packages (from category_encoders) (1.3.1)
Requirement already satisfied: scipy>=1.0.0 in c:\python310\lib\site-packa
ges (from category_encoders) (1.11.3)
Requirement already satisfied: statsmodels>=0.9.0 in c:\python310\lib\site
-packages (from category_encoders) (0.14.3)
Requirement already satisfied: pandas>=1.0.5 in c:\python310\lib\site-pack
ages (from category_encoders) (2.1.1)
Requirement already satisfied: patsy>=0.5.1 in c:\python310\lib\site-packa
ges (from category_encoders) (0.5.6)
Requirement already satisfied: python-dateutil>=2.8.2 in c:\python310\lib
\site-packages (from pandas>=1.0.5->category_encoders) (2.8.2)
Requirement already satisfied: pytz>=2020.1 in c:\python310\lib\site-packa
ges (from pandas>=1.0.5->category_encoders) (2023.3.post1)
Requirement already satisfied: tzdata>=2022.1 in c:\python310\lib\site-pac
kages (from pandas>=1.0.5->category_encoders) (2023.3)
Requirement already satisfied: six in c:\python310\lib\site-packages (from
patsy>=0.5.1->category_encoders) (1.16.0)
Requirement already satisfied: joblib>=1.1.1 in c:\python310\lib\site-pack
ages (from scikit-learn>=0.20.0->category_encoders) (1.3.2)
Requirement already satisfied: threadpoolctl>=2.0.0 in c:\python310\lib\si
te-packages (from scikit-learn>=0.20.0->category_encoders) (3.2.0)
Requirement already satisfied: packaging>=21.3 in c:\python310\lib\site-pa
ckages (from statsmodels>=0.9.0->category_encoders) (23.0)
Downloading category_encoders-2.6.3-py2.py3-none-any.whl (81 kB)
---------------------------------------- 0.0/81.9 kB ? eta -:--:--
----------------------------------- ---- 71.7/81.9 kB 2.0 MB/s eta 0:0
0:01
---------------------------------------- 81.9/81.9 kB 1.5 MB/s eta 0:0
0:00
Installing collected packages: category_encoders
Successfully installed category_encoders-2.6.3

WARNING: Ignoring invalid distribution -eras (c:\python310\lib\site-packag

es)
WARNING: Ignoring invalid distribution -eras (c:\python310\lib\site-packag
es)

[notice] A new release of pip is available: 24.0 -> 24.2

[notice] To update, run: python.exe -m pip install --upgrade pip

In [18]: import category_encoders as ce

encoder = ce.OrdinalEncoder(cols = ['buying', 'maint', 'doors', 'persons',

In [19]: x_train = encoder.fit_transform(x_train)

x_test = encoder.transform(x_test)

localhost:8888/notebooks/BE_PRACTICALS/Ensmble_Learning_ML_5.ipynb 5/7
10/1/24, 11:26 PM Ensmble_Learning_ML_5 - Jupyter Notebook

In [20]: random_forest = RandomForestClassifier(random_state = 0)

​
random_forest.fit(x_train, y_train)
​
y_pred = random_forest.predict(x_test)

In [21]: print(accuracy_score(y_test, y_pred))

0.9479768786127167

In [23]: random_forest_1000 = RandomForestClassifier(n_estimators = 1000, random_sta

In [24]: random_forest_1000.fit(x_train, y_train)

​
y_pred_1000 = random_forest_1000.predict(x_test)

In [25]: print(accuracy_score(y_test, y_pred_1000))

0.953757225433526

In [26]: random_forest_777 = RandomForestClassifier(n_estimators = 777, random_state

​
random_forest_777.fit(x_train, y_train)
​
y_pred_777 = random_forest_777.predict(x_test)
​
print(accuracy_score(y_test, y_pred_777))

0.953757225433526

In [27]: feature_scores = pd.Series(random_forest_1000.feature_importances_, index =

print(feature_scores)

safety 0.293866
persons 0.238120
buying 0.201061
maint 0.125037
lug_boot 0.081547
doors 0.060370
dtype: float64

In [28]: x = df.drop(['class', 'doors'], axis = 1)

​
y = df['class']

In [29]: x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.33, r

In [30]: encoder = ce.OrdinalEncoder(cols = ['buying', 'maint', 'persons', 'lug_boot

In [31]: x_train = encoder.fit_transform(x_train)

x_test = encoder.transform(x_test)

localhost:8888/notebooks/BE_PRACTICALS/Ensmble_Learning_ML_5.ipynb 6/7
10/1/24, 11:26 PM Ensmble_Learning_ML_5 - Jupyter Notebook

In [32]: random_forest = RandomForestClassifier(random_state = 0)

​
random_forest.fit(x_train, y_train)
​
y_pred = random_forest.predict(x_test)

In [33]: print(accuracy_score(y_test, y_pred))

0.9263157894736842

In [34]: from sklearn.metrics import classification_report

​
print(classification_report(y_test, y_pred))

precision recall f1-score support

acc 0.88 0.85 0.86 127

good 0.62 0.56 0.59 18
unacc 0.97 0.97 0.97 399
vgood 0.75 0.81 0.78 26

accuracy 0.93 570

macro avg 0.80 0.80 0.80 570
weighted avg 0.93 0.93 0.93 570

In [ ]: ​

localhost:8888/notebooks/BE_PRACTICALS/Ensmble_Learning_ML_5.ipynb 7/7