Pandas - Ordinal Encoding

Introduction

In this tutorial, we want to encode ordinal categorical variables of a Pandas DataFrame. In order to do this, we use the replace() method of Pandas.

Import Libraries

First, we import the following python modules:

import pandas as pd

Create Pandas DataFrame

Next, we create a Pandas DataFrame with some example data from a dictionary:

data = {
    "language": ["Python", "Python", "Java", "JavaScript"],
    "framework": ["Django", "FastAPI", "Spring", "ReactJS"],
    "users": [20000, 9000, 7000, 5000],
    "popularity": ["High", "High", "Low", "Medium"]
}
df = pd.DataFrame(data)
df

Encode Ordinal Categorical Variables

Now, we would like to convert the categorical values of column "language" into numerical values. We have to consider the rank order of the different elements:

High > Medium > Low

To encode the categorical values, we use the replace() method of Pandas and pass a dictionary with the mapping between categorical and numerical values:

mapping = {
    "Low": 1,
    "Medium": 2,
    "High": 3
}
df["popularity"] = df["popularity"].replace(mapping)
df

Conclusion

Congratulations! Now you are one step closer to become an AI Expert. You have seen that it is very easy to encode ordinal categorical variables of a Pandas DataFrame. We can simply use the replace() method of Pandas. Try it yourself!