Predicting class of flower using Decision tree algorithm

Hi... in this task we gonna determine the class or type of the flower based on its features. The dataset we gonna use is iris dataset. You can use built-in iris dataset or goto UCI repository to download. 

We gonna use the iris built-in dataset on sklearn for this task.

In this task, the machine learning model we gonna built is the decision tree algorithm.

The decision tree is a type of supervised machine learning.where the data is continuously split according to a certain parameter.

But, before that let's quickly recap what is a decision tree.

Decision tree

• Decision tree comes under the category of supervised machine learning.
• Decision tree classification models in the form of a tree structure.
• It breaks down a dataset into smaller and smaller subsets.
• Decision trees are commonly used in decision analysis to help identify a strategy most likely to reach a goal.

We will cover this task on the following steps:

1. Import required libraries

2. Load the iris dataset

3. Prepare the data for the ML model

4. Create the ML model and train it

5. Visualize the ML model 

6. Test the Model 

7. Check the accuracy

1. Import required libraries 

For this task, the libraries we gonna use are NumPy, pandas, Matplotlib and Sklearn.

code 

import pandas as pd

import numpy as np

from sklearn.datasets import load_iris

from sklearn.model_selection import train_test_split

from sklearn.tree import DecisionTreeClassifier

from sklearn import tree

from matplotlib import pyplot as plt

from sklearn.metrics import accuracy_score

2. Load the iris dataset

 

After importing all the libraries now we gonna load the iris dataset to the data frame and observe all its features.

code 

iris=load_iris()

dataframe=pd.DataFrame(iris.data,columns=iris.feature_names)

dataframe.head()

3. Prepare the data for the ML model 

Now it's time to prepare the data for our model. 

To prepare the train data and the test data we gonna use the train_test_split built-in function of the sklearn.

This function will split the data features and data-target into the train data and test data.

This section of the code we split the 120 train dataset and 30 test data.

code   

x=iris.data

y=iris.target

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

print("x train data", x_train.shape)

print("x test data", x_test.shape)

print("y train data", y_train.shape)

print("y test data", y_test.shape)

4. Create the ML model and train it 

Now it's time to create the Machine Learning Model. 

For this task, we gonna use Decision Tree Classifier. 

We will fit the train data and test data to this classifier.

code 

model=DecisionTreeClassifier()

model.fit(x_train,y_train)

print("model is trained")

5. Visualize the ML model 

In this section, we will visualize the decision made by the decision tree to split the data according to a certain parameter.

To do that we gonna use the plot_tree function.

code 

plt.figure(figsize=(20,10))

tree.plot_tree(model, filled=True)

6. Test the Model 

Finally, it's time to check our model. 

To do that we just gonna fit the test data to the model and get the predicted value.

After getting the predicted value we will compare it to the actual value.

code 

prediction=model.predict(x_test)

prediction 

df=pd.DataFrame({'predicted value':prediction,'actual value':y_test})

df['label']=df['predicted value'].replace(dict(enumerate(iris.target_names)))

df.head()

7. Check the accuracy  

The last step let's check the accuracy of our model. 

To do that we gonna use accuracy_score function to to check the accuracy.

code 

score=accuracy_score(prediction,y_test)

print(f"Accuracy score is {score} i.e. {score*100}%")

Video demonstration 

Follow me on: 

Linkedin - https://www.linkedin.com/in/somen-das-6a933115a/  

Instagram - https://www.instagram.com/somen912/?hl=en

Github - https://github.com/somen912/Technical-tasks/blob/main/Task-3%20Decision%20tree.ipynb

And don't forget to subscribe to the blog.

so...

Thanks for your time and stay creative...