Scikit Learn Tutorial – Part I

Installation, Requirements and Building Classification Model

Scikit-learn is one of the most versatile and efficient Machine Learning libraries available across the board. Built on top of other popular libraries such as NumPy, SciPy and Matplotlib, scikit learn contains a lot of powerful tools for machine learning and statistical modelling. No wonder scikit learn is widely used by data scientists, researchers and students alike. Big organizations are using scikit learn to draw insights from the data for making business decisions.

Installing and importing scikit learn

Scikit learn can be installed and imported in the jupyter notebook environment using the following standard commands:

In [5]:

!pip install scikit-learn
import sklearn

That was simple! In the next section, we will discuss the data requirements in scikit learn.

Requirements for working with data in scikit learn 

Before we start training our model using scikit learn, let us understand the following naming conventions in machine learning:

• Features = predictor variables = independent variables
• Target variable = dependent variable = response variable
• Samples=records=instances

The scikit learn library has the following requirements for the data before it can be used to train a model:

1. Features and response should be separate objects
2. Features and response should be numeric
3. Features and response should be NumPy arrays of compatible sizes (in terms of rows and columns)

Next, we will see an example of a dataset which meets the above requirements to be used in scikit learn.