Visualization and Case study

Uber dataset (from Kaggle.com)

In [1]:
import numpy as np
import pandas as pd
In [2]:
iris_df = pd.read_csv('iris.csv')
iris_df.head()
Out[2]:
Id SepalLengthCm SepalWidthCm PetalLengthCm PetalWidthCm Species
0 1 6.5 3.2 5.1 2.0 Iris-virginica
1 2 6.1 2.8 4.0 1.3 Iris-versicolor
2 3 5.1 3.5 1.4 0.3 Iris-setosa
3 4 6.4 3.1 5.5 1.8 Iris-virginica
4 5 6.7 3.1 4.7 1.5 Iris-versicolor
In [3]:
iris_df.tail()
Out[3]:
Id SepalLengthCm SepalWidthCm PetalLengthCm PetalWidthCm Species
145 146 4.6 3.6 1.0 0.2 Iris-setosa
146 147 5.0 3.3 1.4 0.2 Iris-setosa
147 148 6.0 3.0 4.8 1.8 Iris-virginica
148 149 6.7 3.1 5.6 2.4 Iris-virginica
149 150 5.3 3.7 1.5 0.2 Iris-setosa
In [4]:
# Remove the id columns as it doesn't contribute towards the data
iris_df.drop('Id', axis=1, inplace=True)
iris_df
Out[4]:
SepalLengthCm SepalWidthCm PetalLengthCm PetalWidthCm Species
0 6.5 3.2 5.1 2.0 Iris-virginica
1 6.1 2.8 4.0 1.3 Iris-versicolor
2 5.1 3.5 1.4 0.3 Iris-setosa
3 6.4 3.1 5.5 1.8 Iris-virginica
4 6.7 3.1 4.7 1.5 Iris-versicolor
... ... ... ... ... ...
145 4.6 3.6 1.0 0.2 Iris-setosa
146 5.0 3.3 1.4 0.2 Iris-setosa
147 6.0 3.0 4.8 1.8 Iris-virginica
148 6.7 3.1 5.6 2.4 Iris-virginica
149 5.3 3.7 1.5 0.2 Iris-setosa

150 rows × 5 columns

Univariate Analysis

In [5]:
iris_df.shape
Out[5]:
(150, 5)
In [6]:
iris_df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 150 entries, 0 to 149
Data columns (total 5 columns):
 #   Column         Non-Null Count  Dtype  
---  ------         --------------  -----  
 0   SepalLengthCm  150 non-null    float64
 1   SepalWidthCm   150 non-null    float64
 2   PetalLengthCm  150 non-null    float64
 3   PetalWidthCm   150 non-null    float64
 4   Species        150 non-null    object 
dtypes: float64(4), object(1)
memory usage: 6.0+ KB
In [7]:
iris_df.describe().T
Out[7]:
count mean std min 25% 50% 75% max
SepalLengthCm 150.0 5.843333 0.828066 4.3 5.1 5.80 6.4 7.9
SepalWidthCm 150.0 3.054000 0.433594 2.0 2.8 3.00 3.3 4.4
PetalLengthCm 150.0 3.758667 1.764420 1.0 1.6 4.35 5.1 6.9
PetalWidthCm 150.0 1.198667 0.763161 0.1 0.3 1.30 1.8 2.5
In [8]:
iris_df.isnull().sum()

# Purpose column has a lot of Null values
Out[8]:
SepalLengthCm    0
SepalWidthCm     0
PetalLengthCm    0
PetalWidthCm     0
Species          0
dtype: int64

Mapping labels to numerical classes

In [9]:
iris_df['Species'].unique()
Out[9]:
array(['Iris-virginica', 'Iris-versicolor', 'Iris-setosa'], dtype=object)
In [10]:
iris_df['Species'] = iris_df['Species'].map({ 'Iris-virginica': 1, 'Iris-versicolor': 2, 'Iris-setosa': 3 })
iris_df
Out[10]:
SepalLengthCm SepalWidthCm PetalLengthCm PetalWidthCm Species
0 6.5 3.2 5.1 2.0 1
1 6.1 2.8 4.0 1.3 2
2 5.1 3.5 1.4 0.3 3
3 6.4 3.1 5.5 1.8 1
4 6.7 3.1 4.7 1.5 2
... ... ... ... ... ...
145 4.6 3.6 1.0 0.2 3
146 5.0 3.3 1.4 0.2 3
147 6.0 3.0 4.8 1.8 1
148 6.7 3.1 5.6 2.4 1
149 5.3 3.7 1.5 0.2 3

150 rows × 5 columns

Checking the skewness

In [11]:
import matplotlib.pyplot as plt

iris_df.hist(figsize=(12,12))
plt.show();

Multivariate analysis

Pearson's correlation coefficient

In [12]:
iris_df_corr = iris_df.corr('pearson')['Species']
iris_df_corr_valuable = iris_df_corr[ (iris_df_corr >= 0.3) | (iris_df_corr <= -0.2)]
iris_df_corr_valuable
Out[12]:
SepalLengthCm   -0.782561
SepalWidthCm     0.419446
PetalLengthCm   -0.949043
PetalWidthCm    -0.956464
Species          1.000000
Name: Species, dtype: float64

Pairplot the data to visualize the relatioship

In [13]:
import seaborn as sns

sns.pairplot(iris_df, diag_kind='kde')
plt.show();
In [14]:
for colname in iris_df.columns:
    iris_df[[colname]].boxplot();
    plt.show();