Feature Engineering for Existing Features.png

Feature engineering is a crucial stage in any machine learning project. It allows you to use data to define features that enable machine learning algorithms to work properly.

In this course, you will learn the techniques that will help you create new features from existing features. You’ll start by diving into label encoding which is crucial for converting categorical features into numerical. You’ll also learn about other various types of encoding such as: one-hot, count, and mean, all of which are important for feature engineering. 

In the remaining chapters, you’ll learn about feature interaction and datetime features. In all, this course will show you the many different ways you can create features from existing ones.

Feature Engineering for Machine Learning

DateTime is an essential feature because some behaviors have time patterns. In this lesson, we would introduce two ways to deal with time features.

>**Notice**: In the real world, the time is recorded as the UNIX timestamp. Therefore, all the demos in this lesson will use the UNIX timestamp as the raw DateTime format.

# Extract month, week, and hour with `pandas.`

Users' behaviors have time patterns. For example, there is a significant difference between what users buy in winter and summer. Users' preferences vary from time of day to time of day. 

Here we show how to extract the month, week, and hour with `pandas` from UNIX timestamp:

```python
df["col1"] = pd.to_datetime(df['col1'], unit='s')
df["month"] = df["col1"].dt.month
```

In this lesson, we introduce two ways to deal with time features.

__default

Datetime Features