Model Evaluation - lesson 5 | HackIndia Skills

Model Evaluation - lesson 5 | HackIndia Skills | HackIndia

Example

Majority baseline code

This example uses scikit-learn. It predicts the most common class from the training labels, then evaluates on validation labels.

Use this for a first baseline in a binary or multi-class classification project.

from sklearn.dummy import DummyClassifier
from sklearn.metrics import confusion_matrix, classification_report

# X_train, X_val, y_train, y_val should already come from your clean split

baseline = DummyClassifier(strategy="most_frequent")
baseline.fit(X_train, y_train)

base_pred = baseline.predict(X_val)

print("Baseline confusion matrix")
print(confusion_matrix(y_val, base_pred))

print("Baseline metrics")
print(classification_report(y_val, base_pred, zero_division=0))

Example

How to write it

Your hackathon project does not need a long research paper. Add a short baseline line in the evaluation section.

Keep it direct, like this:

Baseline: DummyClassifier with strategy='most_frequent'.
It always predicts 'not_fraud', the majority class in the training set.
Validation split: 80/20 stratified split, fixed random_state=42.
Metric: recall for 'fraud', because missing fraud is costly.
Baseline recall: 0.00. Our model recall: 0.71.
Limitation: The validation set is small, so this result may change on new UPI-like transaction data.

Beat a real baseline

One idea

What counts as baseline

Bad baseline choices

Pick one fairly

Majority baseline code

Use AI carefully

How to write it

Do this now

Quick check