Lære Testing ML Models

Sveip for å vise menyen

Automated testing is a cornerstone of robust machine learning (ML) deployment pipelines. By systematically verifying each component of your ML workflow, you can catch errors early, prevent regressions, and ensure that your models perform reliably in production. Testing strategies for ML models typically include both unit tests and integration tests.

Unit tests focus on small, isolated pieces of code—such as data preprocessing functions or feature engineering steps—ensuring that each performs as expected;
Integration tests, on the other hand, validate that multiple components work together correctly, such as checking that a trained model produces inference outputs with the correct shape and type when given new data.


              123456789101112131415161718192021
            
import unittest
import numpy as np
from sklearn.linear_model import LogisticRegression

class TestModelPredictionShape(unittest.TestCase):
    def test_prediction_output_shape(self):
        # Simulate training data
        X_train = np.random.rand(10, 3)
        y_train = np.random.randint(0, 2, 10)
        # Train a simple model
        model = LogisticRegression()
        model.fit(X_train, y_train)
        # Simulate new input data
        X_new = np.random.rand(5, 3)
        # Get predictions
        predictions = model.predict(X_new)
        # Assert that output shape matches expected
        self.assertEqual(predictions.shape, (5,))

if __name__ == "__main__":
    unittest.main(argv=[''], exit=False)

Alt var klart?

Takk for tilbakemeldingene dine!

Seksjon 1. Kapittel 3

Spør AI

Spør om hva du vil, eller prøv ett av de foreslåtte spørsmålene for å starte chatten vår

Seksjon 1. Kapittel 3