Setup and Your First Model

scikit-learn is Python's most widely used library for classical machine learning: classification, regression, clustering, and everything around evaluating and tuning those models. Unlike this course's quantum computing or game-development siblings, scikit-learn (along with numpy and pandas) actually runs directly in this course's browser sandbox, every lesson here is genuinely runnable, no local install required.

What "machine learning" actually means

A traditional program is a set of rules a human writes by hand: "if the email contains these words, mark it as spam." A machine learning model instead learns those rules from examples: show it thousands of emails already labeled spam/not-spam, and it figures out the patterns itself. scikit-learn's entire API is built around two verbs that capture this idea:

• fit(X, y), learn patterns from labeled training data (X = inputs, y = correct answers).
• predict(X), apply what was learned to new, unseen inputs.

Your first model

from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.tree import DecisionTreeClassifier

iris = load_iris()
X_train, X_test, y_train, y_test = train_test_split(
    iris.data, iris.target, test_size=0.2, random_state=42
)

model = DecisionTreeClassifier(random_state=42)
model.fit(X_train, y_train)

predictions = model.predict(X_test)
print('Predicted:', predictions[:10])
print('Actual:   ', y_test[:10])

• load_iris() loads a small, classic dataset (bundled with scikit-learn, no download needed) of 150 flowers, each described by 4 measurements (.data) and labeled with one of 3 species (.target).
• train_test_split(...) splits the data into a chunk for training and a chunk held back for testing, covered in depth next lesson.
• DecisionTreeClassifier() creates an untrained model, a decision tree specifically, but every scikit-learn model shares this same fit/predict shape regardless of what's happening internally.
• model.fit(X_train, y_train) is where the actual learning happens, before this line, model knows nothing about flowers.
• model.predict(X_test) asks the now-trained model to guess the species of flowers it has never seen, comparing its guesses to the real answers is exactly how you'll judge whether it learned anything useful.