Machine Learning with Python – Introduction and First Model

Key Machine Learning Concepts

Supervised learning: learn from labelled examples (classification, regression). Unsupervised learning: find patterns in unlabelled data (clustering, dimensionality reduction). Reinforcement learning: learn from rewards and penalties.

# The ML workflow:
# 1. Collect and prepare data
# 2. Split into training and test sets
# 3. Choose and train a model
# 4. Evaluate on test set
# 5. Tune and improve
# 6. Deploy

print("Data → Model → Predictions")

scikit-learn – Python's ML Toolkit

scikit-learn provides consistent APIs for 50+ ML algorithms, plus data preprocessing, model evaluation, and pipelines.

# pip install scikit-learn
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.metrics import accuracy_score

# Load example dataset
iris = load_iris()
X, y = iris.data, iris.target

print(f"Samples: {len(X)}, Features: {X.shape[1]}")
print(f"Classes: {iris.target_names}")

Training Your First Model

Split data, train a classifier, and evaluate it.

from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.tree import DecisionTreeClassifier
from sklearn.metrics import accuracy_score, classification_report

iris = load_iris()
X, y = iris.data, iris.target

# Split: 80% train, 20% test
X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.2, random_state=42
)

# Train
model = DecisionTreeClassifier(max_depth=3, random_state=42)
model.fit(X_train, y_train)

# Evaluate
y_pred = model.predict(X_test)
print(f"Accuracy: {accuracy_score(y_test, y_pred):.2%}")

Linear Regression Example

Predict continuous values (price, temperature, salary) with linear regression.

import numpy as np
from sklearn.linear_model import LinearRegression
from sklearn.model_selection import train_test_split
from sklearn.metrics import mean_squared_error, r2_score

# Synthetic house price data
np.random.seed(42)
sqft = np.random.randint(500, 3000, 200).reshape(-1, 1)
price = sqft * 250 + np.random.normal(0, 15000, (200,1))

X_train, X_test, y_train, y_test = train_test_split(sqft, price, test_size=0.2)

model = LinearRegression()
model.fit(X_train, y_train)

y_pred = model.predict(X_test)
print(f"R² Score: {r2_score(y_test, y_pred):.3f}")
print(f"Predict 2000 sqft: ${model.predict([[2000]])[0][0]:,.0f}")

The Machine Learning Workflow in Python

Machine learning finds patterns in data instead of following hand-coded rules. In Python, scikit-learn gives every model the same simple interface, so the workflow matters more than any one algorithm.

from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier

# 1. split — NEVER test on data you trained on
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)

# 2. fit (learn) then 3. predict, 4. evaluate
model = RandomForestClassifier().fit(X_train, y_train)
model.score(X_test, y_test)      # accuracy on UNSEEN data

The cardinal rule — train/test split: always evaluate on data the model never saw during training. A model that scores 99% on its training data but fails on new data is overfitting — memorizing, not learning. The held-out test set is your honest estimate of real-world performance. Reality check: most ML work is data preparation — cleaning, encoding, and scaling features — not choosing algorithms. Every scikit-learn model follows the same .fit() / .predict() pattern, so learning the workflow transfers across all of them.

Interview Questions on Machine Learning

• What is machine learning and how does it differ from traditional programming?
• What is the difference between supervised and unsupervised learning?
• What is the purpose of splitting data into training and test sets?
• What is overfitting and how do you detect it?
• What does scikit-learn’s fit/predict API do?
• What is the difference between classification and regression?
• What are features and labels?

Related Topics