Build Your First Machine Learning Model in Python (20-Minute Tutorial)

What You’ll Need

• Python 3.8+ installed
• Virtual environment (venv, Conda, etc.)
• Familiarity with Python basics
• Libraries: scikit-learn, pandas, numpy, matplotlib

Step 1 — Load Your Dataset

We’ll use the built-in Iris dataset:

from sklearn.datasets import load_iris
import pandas as pd

iris = load_iris()
X = pd.DataFrame(iris.data, columns=iris.feature_names)
y = pd.Series(iris.target, name='species')

print(X.head())
print(y.value_counts())

Step 2 — Split into Train & Test

Keep an 80/20 split to evaluate on unseen data:

from sklearn.model_selection import train_test_split

X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.2, random_state=42
)

Step 3 — Train a Simple Classifier

Using Logistic Regression:

from sklearn.linear_model import LogisticRegression

model = LogisticRegression(max_iter=200)
model.fit(X_train, y_train)

Step 4 — Evaluate Model Performance

Check accuracy and visualize a confusion matrix:

from sklearn.metrics import accuracy_score, confusion_matrix, classification_report
import matplotlib.pyplot as plt

y_pred = model.predict(X_test)

print("Accuracy:", accuracy_score(y_test, y_pred))
print(classification_report(y_test, y_pred, target_names=iris.target_names))

cm = confusion_matrix(y_test, y_pred)
plt.imshow(cm, cmap='Blues', interpolation='nearest')
plt.title("Confusion Matrix")
plt.xlabel("Predicted")
plt.ylabel("Actual")
plt.colorbar()
plt.show()

Step 5 — Make a New Prediction

Use your model on a custom example:

sample = [[5.1, 3.5, 1.4, 0.2]]
pred_class = iris.target_names[model.predict(sample)[0]]
print("Predicted species:", pred_class)

Step 6 — Deploy as a Minimal API (Bonus)

Serve predictions with Flask:

from flask import Flask, request, jsonify

app = Flask(__name__)

@app.route('/predict', methods=['POST'])
def predict():
    data = request.json['features']
    pred = model.predict([data])[0]
    return jsonify({"species": iris.target_names[pred]})

Next: containerize with Docker or deploy on AWS Lambda.

Common Gotchas & Tips

• Scale features with StandardScaler().
• Prevent overfitting by monitoring train/test gap.
• Use GridSearchCV for hyperparameter tuning.

What You’ve Achieved

1. Loaded & explored data
2. Trained & evaluated a model
3. Visualized performance
4. Served predictions via API

Level-Up with the Full Course

Master advanced ML techniques, live projects, & career coaching.