Code360 powered by Coding Ninjas X Code360 powered by Coding Ninjas X
Table of contents
Frequently Asked Questions
What do we mean by Logistic regression?
What do we mean by Normal distribution in Logistic regression?
What are the features of Normal distribution?
What is a Histogram?
What do we mean by Centralisation of Random Variables?
Recommended Readings
Also, check out
Last Updated: Mar 27, 2024

Applying Logistic Regression on Iris dataset

Author Rajkeshav
0 upvote


There are several machine learning models present today. Different types of models, like different people, have their specialties and fields where they excel more than others. Here in this article, we are going to employ the services of the Logistic Regression Model on the Iris Dataset to predict the species of the flower. 

Logistic Regression is used to predict a dependent variable, given a set of independent variables, such that the dependent variable is categorical.

The Logistic function is an S-shaped curve that can take any real-valued number and map it into a value between 0 and 1. 

Linear Model v/s Logistic Model


Sigmoid function: 11+e-x , x is the independent variable.


Let's import some essential modules. 

import pandas as pd
# used to read the data set
import numpy as np
# used to do some operations with the arrays
import os
# used handle some files
import matplotlib.pyplot as plt
# used to visualize the data using graphs
import seaborn as sns
# plotting the chart in a single line


Let's load the dataset (see Pandas).

df = pd.read_csv("Iris.csv")


The data set will be read and stored in the data frame format.

Let’s display the first five rows of the data set.





We have four features, SepalLength, SepalWidth, PetalLength, and PetalWidth. The last feature, 'Species,' is the target feature that will be predicted.

Feature 'Id' is a definite feature that does not affect the target feature. So let's drop that feature.


df = df.drop(columns = ['Id'])




Let’s have the information about the data type of the data set.

SepalLength, SepalWidth, PetalLength, and PetalWidth have float data types. 'Species' has an object data type. Let's check the number of samples of each class in Species.






In each class, we have fifty samples.

While training the model, we must remove all the null values with other deals like mean, median, mode, etc. To check whether the data set contains the null values, We write.




It will display the number of null values in each column.




There are no null or nan values in the datasets.

Now, We will visualize the data in the form of graphs. First, let's display some basic charts. For each column, let us create a histogram.




Sepal Histogram

          Blue - SepalLenth

          Orange- SepalWidth




Petal Histogram


        Blue- PetalLength

        Orange- PetalWidth


We can observe that the histogram plot of all the features follows the normal distribution. 

PDF of the normal distribution with mean μ and variance σ2 is given by:


 Normal Distribution Formula


If the size of the data set is large enough, the normal distribution is a good approximation.

A correlation matrix shows the correlation between the two variables. The value is in the range of -1 to +1. If two variables have a correlation much nearer to +1 or -1, we can neglect one variable.

So we are going to construct the matrix from the input attributes.

If you want to learn more about covariance and correlation check out Covariance and Correlation






Here, the correlation coefficient between PetalLength and PetalWidth is 0.962757, which is much closer to 1, so we can remove any of the two variables. A correlation coefficient is a convenient tool when we have lots of features, and we can neglect multiple features by observing the correlation coefficient matrix. 


In Machine learning, we usually deal with the dataset which contains multiple labels in columns. These labels can be in the form of words or numbers. Label Encoding refers to converting the labels into the numeric form into the machine-readable form.

The output class is in the categorical form in this data set, and we need to convert it into the numeric format. So We will use Label Encoder.

from sklearn.preprocessing import LabelEncoder
le = LabelEncoder()
df['Species'] = le.fit_transform(df['Species'])




After preprocessing all the data sets, we need to train and test our model. So, let's import the model.


from sklearn.model_selection import train_test_split


We need to separate the classes for input and output attributes. Let's drop the columns ‘Species’ and store the remaining input attributes to the variable X. The output attribute, i.e., ‘Species,’ is taken in the variable Y.


X = df.drop(columns = ['Species'])
Y = df['Species']
X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size = 0.25)


Now let's import a basic classification model called logistic Regression from SKlearn.


from sklearn.linear_model import LogisticRegression
model = LogisticRegression()


Now, let's train the model., Y_train)


Let's print the metric to get the performance.


print("Accuracy: ", model.score(X_test, Y_test) * 100)


We are getting an accuracy of 92%, which is very decent. The accuracy increases with the increase in the number of data points.
Check out this problem - Largest Rectangle in Histogram

Get the tech career you deserve, faster!
Connect with our expert counsellors to understand how to hack your way to success
User rating 4.7/5
1:1 doubt support
95% placement record
Akash Pal
Senior Software Engineer
326% Hike After Job Bootcamp
Himanshu Gusain
Programmer Analyst
32 LPA After Job Bootcamp
After Job

Frequently Asked Questions

What do we mean by Logistic regression?

Logistic Regression is used to predict a dependent variable, given a set of independent variables, such that the dependent variable is categorical.

What do we mean by Normal distribution in Logistic regression?

The normal distribution is a bell-shaped curve, and many continuous data in nature follow the bell-shaped curve when compiled and graphed. 

What are the features of Normal distribution?

Following are the properties of Normal Distribution:
1) The mean of the data points lies where the curve has the maximum value.
2) The area under the Normal distribution curve is always unity divided equally into two halves in both directions. 
3) The area under the curve between two points is the definite integration of the probability density function of the normal distribution with limits at those points.

What is a Histogram?

A Histogram is a bar graph representation of the data points that are used to compare one feature with others.

What do we mean by Centralisation of Random Variables?

While dealing with the random variables, one might get distracted by the mean as it could be anywhere. So, the efficient idea is to make the mean zero. This is done by creating a new random variable from the old one such that the mean is always zero. This is called Centralisation.


We discussed the implementation of Logistic Regression on the Iris Dataset in the above blogs. One can argue that there may be more optimal methods for classification in the iris dataset; one of them is K-means Algorithm. Feel free to try that out and compare the results. If you want to Machine Learning a bit more, check out the Guided path for Machine Learning on the Coding Ninjas Studio.

Recommended Readings

Also, check out

Happy Coding!!

Live masterclass