Search⌘ K
AI Features
Home
Courses
Mastering Hyperparameter Optimization for Machine Learning

Data Preparation

Explore how to prepare a dataset for machine learning by loading, cleaning, handling missing values, and feature engineering. Understand how to split data into features and target variables, scale numerical data, and convert categorical data into numerical formats. This lesson equips you with foundational data preparation skills essential for effective hyperparameter optimization.

We'll cover the following...

How to prepare the dataset

In this lesson, we’ll explore how to load the dataset and prepare it for training an ML model using its default hyperparameters. We’ll then apply the hyperparameter techniques to improve performance.

What will we learn?

We’ll learn to:

  • Load the dataset.

  • Clean the dataset.

  • Perform feature engineering techniques to preprocess the dataset.

Import important packages

First, we’ll import important Python packages that will do the following tasks:

  • Load the dataset.

  • Clean the dataset.

  • Process the dataset using feature engineering techniques.

Python 3.8
# import important modules
import numpy as np
import pandas as pd
# sklearn modules
from sklearn.preprocessing import MinMaxScaler
import warnings
warnings.filterwarnings("ignore")
# seeding
np.random.seed(123)

Load the dataset

We’ll use pandas to load the dataset from the data folder. The name of the dataset is loan_data.csv.

Python 3.8
# load data
data_path = "loan_data.csv"
data = pd.read_csv(data_path)

Let’s see the first five rows of the dataset using the head() method from pandas.

Python 3.8
# show the top five rows of data
data.head()

The above code will show the first five rows of the dataset.

Data from the first five rows and first ten columns
Data from the first five rows and first ten columns

As we can see, the dataset has 13 columns:

  • Loan_ID: ...