Key Concepts

Datasets

A dataset is a structured collection of examples. Each row is one instance (also called a sample or observation).

The Iris dataset

The Iris dataset, introduced by British statistician and biologist Ronald A. Fisher in 1936, is one of the most iconic datasets in the history of machine learning and statistics. It contains measurements of 150 iris flowers, evenly split across three species — Iris setosa, Iris versicolor, and Iris virginica — with each flower described by four features: sepal length, sepal width, petal length, and petal width, all measured in centimetres. Its appeal lies in its simplicity and structure: one class (setosa) is linearly separable from the other two, while the remaining pair overlaps slightly, making it an ideal sandbox for exploring both simple and more nuanced classification algorithms.

Code

import pandas as pd
from sklearn.datasets import load_iris
import matplotlib.pyplot as plt

iris = load_iris(as_frame=True)
df = iris.frame
df["species"] = df["target"].map({0: "setosa", 1: "versicolor", 2: "virginica"})
df = df.drop(columns=["target"])

df.head(6)

	sepal length (cm)	sepal width (cm)	petal length (cm)	petal width (cm)	species
0	5.1	3.5	1.4	0.2	setosa
1	4.9	3.0	1.4	0.2	setosa
2	4.7	3.2	1.3	0.2	setosa
3	4.6	3.1	1.5	0.2	setosa
4	5.0	3.6	1.4	0.2	setosa
5	5.4	3.9	1.7	0.4	setosa

The first few rows of the Iris dataset.

Features and Labels

Features (\(X\)): The input variables, characteristics, or attributes we use to make predictions.

Labels (\(y\)): The target variable we want the model to predict.

Term	Meaning	Example (Iris)
Feature	An input variable used to make predictions	petal length, sepal width
Label	The output the model tries to predict	species name
Instance	One row — one complete example	a single flower measurement

Code

colours = {"setosa": "#4C72B0", "versicolor": "#55A868", "virginica": "#C44E52"}

fig, ax = plt.subplots(figsize=(7, 4))
for species, group in df.groupby("species"):
    ax.scatter(group["petal length (cm)"], group["petal width (cm)"],
               label=species, color=colours[species], alpha=0.8,
               edgecolors="white", s=60)
ax.set_xlabel("Petal length (cm)  ← feature")
ax.set_ylabel("Petal width (cm)   ← feature")
ax.set_title("Iris dataset — features coloured by label")
ax.legend(title="Label (species)")
plt.tight_layout()
plt.show()

Two features coloured by label (species)

Training vs. Testing

We split data into two sets:

Training set — the model sees this and learns from it
Test set — never seen during training; used to estimate real-world performance

Code

from sklearn.model_selection import train_test_split

X = df[["petal length (cm)", "petal width (cm)"]]
y = df["species"]
X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.2, random_state=42
)

print(f"Training samples : {len(X_train)}  ({len(X_train)/len(df):.0%})")
print(f"Test samples     : {len(X_test)}  ({len(X_test)/len(df):.0%})")

fig, axes = plt.subplots(1, 2, figsize=(10, 4), sharey=True)
for ax, (split_X, split_y, title) in zip(axes, [
    (X_train, y_train, f"Training set  (n={len(X_train)})"),
    (X_test,  y_test,  f"Test set  (n={len(X_test)})")
]):
    for species, colour in colours.items():
        mask = split_y == species
        ax.scatter(split_X.loc[mask, "petal length (cm)"],
                   split_X.loc[mask, "petal width (cm)"],
                   label=species, color=colour, alpha=0.8,
                   edgecolors="white", s=60)
    ax.set_title(title)
    ax.set_xlabel("Petal length (cm)")
axes[0].set_ylabel("Petal width (cm)")
axes[1].legend(title="Species")
plt.tight_layout()
plt.show()

Training samples : 120  (80%)
Test samples     : 30  (20%)

An 80/20 train/test split on the Iris dataset

Warning

Data leakage — accidentally letting the model see test data during training — is one of the most common ML mistakes. It produces falsely optimistic performance estimates.