A Beginner Tutorial For ML Decision Tree Model Using Scikit And Panda

Prerequisite

Python, Scikit and Panda installed in your laptop. It’s better to install conda as it has all the required libraries. Install Jupyter too. It really helps in python coding.

Panda

Panda is a popular python library to explore and manipulate data.

Scikit

Scikit is popular machine learning framework in python.

Regression

Regression is process to find relation between one variable and several dependent variable. There are many regression techniques like linear regression, simple regression ordinary least squares to name a few.

Decision Tree Regression

Suppose you have the following data with you about the number of a bathroom in a house and it’s price:

 

Number of Bathroom Price
1 10000
2 10000
3 30000

You might infer from the data above that whenever number of bathrooms in house is less than three, price is 10000 else it is 30000. Same inference can be put in the following way.

Screen Shot 2017-11-19 at 3.30.52 AM

This is an example of decision tree, albeit very crude of level one. Here, we have only two leaf. So, the lack of data makes us think that if a house has 7 bathrooms, it will still have 30000 as price. Now, you can think that scikit models can go upto level 10 which will have around 1000 leaf and that model will be more accurate than this. We have used Decision Tree Regression to predict the pricing of House in Melbourne.

Dataset

Please download the dataset from here.

Example

In the example shown below, comments are added at each step. Please go through the code once and make your first ML model. This example has been done in jupyter notebook so ignore comments like #In[6]

 

Find how to implement ml model in java here.

You can also see tutorial of DanB at kaggle here.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Blog at WordPress.com.

Up ↑

%d bloggers like this: