Lesson menuPython TutorialHideShow

Python Tutorial

Categorical Data

Turn text categories into numbers that models can use — without inventing a false ordering.

← Grid SearchK-means →

Why Encoding Is Needed

Most models do arithmetic on their inputs, so a column like Color = ["red", "green", "blue"] must become numeric. The trick is to encode categories without implying an order that does not exist — "blue" is not greater than "red".

One-Hot Encoding

One-hot encoding creates a separate 0/1 column per category. It is the safe default for nominal (unordered) features.

import pandas as pd

df = pd.DataFrame({
    "Color": ["red", "green", "blue", "green"],
    "Price": [10, 15, 20, 12],
})

encoded = pd.get_dummies(df, columns=["Color"], drop_first=False)
print(encoded)
#    Price  Color_blue  Color_green  Color_red
# 0     10       False        False       True
# ...

drop_first=True removes one column to avoid the "dummy variable trap" (perfect collinearity) in linear models. Tree models do not need this.

Ordinal Encoding

When categories have a genuine order — ["low", "medium", "high"] — map them to ranked integers so the model can use the ordering.

from sklearn.preprocessing import OrdinalEncoder

sizes = [["low"], ["high"], ["medium"], ["low"]]
enc = OrdinalEncoder(categories=[["low", "medium", "high"]])
print(enc.fit_transform(sizes).ravel())   # [0. 2. 1. 0.]

Never apply ordinal encoding to unordered categories. Telling the model red=0, green=1, blue=2 invents a ranking that will distort distance and linear models.

Encoding in a scikit-learn Pipeline

ColumnTransformer applies different encoders to different columns and remembers the categories learned on the training set — essential for correctly transforming new data.

from sklearn.compose import ColumnTransformer
from sklearn.preprocessing import OneHotEncoder, StandardScaler
from sklearn.pipeline import Pipeline
from sklearn.linear_model import LinearRegression

pre = ColumnTransformer([
    ("cat", OneHotEncoder(handle_unknown="ignore"), ["Color"]),
    ("num", StandardScaler(), ["Price"]),
])

model = Pipeline([("pre", pre), ("reg", LinearRegression())])
model.fit(df[["Color", "Price"]], [1, 2, 3, 2])

handle_unknown="ignore" keeps the pipeline from crashing when the test set contains a category the training set never saw.

High-Cardinality Features

One-hot encoding a column with thousands of categories (zip codes, product IDs) explodes the feature count. Alternatives:

  • Target / mean encoding: replace each category with the mean target value (fit inside CV to avoid leakage).
  • Frequency encoding: replace each category with how often it appears.
  • Hashing / embeddings: map categories into a fixed-size space.

Best Practices

  • One-hot for nominal features; ordinal only when a real order exists.
  • Fit the encoder on training data and reuse it on test/new data.
  • Use handle_unknown="ignore" so unseen categories do not break inference.
  • Prefer target/frequency encoding for very high-cardinality columns.

Try It Yourself

Exercise 1: One-hot encode a column of colours with pandas.

Show solution
import pandas as pd
df = pd.DataFrame({"color": ["red", "blue", "red"]})
print(pd.get_dummies(df, columns=["color"]))

Exercise 2: Would you ordinal-encode ["small", "medium", "large"] or one-hot encode it? Why?

Show solution

Ordinal encoding — the categories have a genuine order (small < medium < large) that the model can use.

📘 Real-World Deep Dive

Knowing <strong>ML Categorical Data (scikit-learn)</strong> well is what turns scikit-learn from a curiosity into a daily tool — you'll reach for it in nearly every real project.

Real-Life Scenario

An end-to-end usage of ML Categorical Data that you'd actually see in a data pipeline or analytics notebook.

Real-Life Example

import numpy as np
from sklearn.preprocessing import OneHotEncoder
X = np.array([["red"], ["blue"], ["red"]])
print(OneHotEncoder().fit_transform(X).toarray())

Expected Output

(see source)

Common mistakes

  • fit expects numeric arrays; OneHotEncoder / LabelEncoder are easy to forget for categorical features.
  • Calling predict on a model trained on unscaled data and then scaling inputs at inference time silently degrades accuracy.
  • train_test_split(X, y) requires both arrays; train_test_split(X) for unsupervised learning slips past static checkers.
  • Treating ML Categorical Data as a black box without reading the docs — the API has subtle defaults that bite when you scale.

🚀 Performance & Best Practices

  • Wrap preprocessing + estimator in a Pipeline so fit / predict stay reproducible.
  • Use joblib / pickle for serialising models, not the entire Python state.
  • For > 100 k rows, switch to HistGradientBoostingClassifier or cuML.
  • When working with scikit-learn, prefer vectorised / batched operations over Python loops.

🧪 Try It Yourself

  1. Reproduce the snippet on a representative slice of your own data.
  2. Profile the snippet with cProfile or timeit and find the single biggest improvement.
  3. Generalise the snippet into a small, reusable function you can drop into future projects.

FAQ: Categorical Data

Common questions about this page.

What is Categorical Data?

Categorical Data is a Machine Learning lesson that explains categorical data in Python. Turn text categories into numbers that models can use — without inventing a false ordering. Copy the samples and run them in the Python editor. It is written for beginners who want a clear definition and working examples.

Should I run categorical data examples locally for better learning?

Yes. Use the browser editor on StudyGrid for a quick check, then Download the example and run it on your computer. Local runs show real errors and the real toolchain, which is one of the fastest ways to learn categorical data in this Machine Learning Python lesson (Categorical Data).

How do I use categorical data in Python?

To use categorical data in Python, follow the examples on this StudyGrid page. Copy a snippet, run it in the browser, then Download and run it locally for better learning. Change the values and compare the output.

What is the syntax of categorical data?

This Categorical Data tutorial shows categorical data syntax with short Python examples. Use the code blocks in this lesson for the exact statements, then try them in your editor.

Categorical Data example for beginners

Yes. This page includes a beginner categorical data example you can copy and run. It is designed for searches such as "categorical data for beginners", "categorical data example", and "how to use categorical data".

What are common mistakes with categorical data?

Common categorical data mistakes include wrong syntax, mixing types, and skipping practice. Work through this Machine Learning chapter in order, run every example, and check the output before moving on.

Why should I learn categorical data?

Categorical Data is used in real Python work. Learning categorical data helps you write clearer programs and continue the Machine Learning tutorial on StudyGrid.

Is Categorical Data free to learn online?

Yes. You can learn categorical data free on StudyGrid (studygrid.in). This chapter is part of the Machine Learning path and includes examples, syntax, and next-step links.