Intermediate level · Module 6: Intermediate Machine Learning Project · Lesson 2 of 3
Building and Evaluating
Build the pipeline, tune it, segment agents and evaluate everything against the brief.
In this lesson you will
- Apply pipelines, cross-validation and tuning to the project
- Evaluate against the success criteria
Key concepts
pipelinetuning logtest evaluationsegment-wise performance
The lesson
Use the techniques from this level: engineered features in a pipeline, cross-validated comparison with baselines, a documented search and a single final test evaluation.
Check performance by segment and region; an overall success can hide failure for a group of agents.
Compare the results with the success criteria in the brief and say plainly which were met.
Step by step
- Build and compare candidate pipelines.
- Tune the chosen one and evaluate on the test set.
- Produce the segmentation.
- Compare results with the success criteria.
Practice activity
Complete the modelling and segmentation work for the project.
Expected result: A tuned model and segmentation with results checked against the brief.
Check your understanding
-
Why check performance by segment?
Show answer
Good overall results can hide poor results for some groups.
-
When is the test set used?
Show answer
Once, for the final evaluation.
Taught live by an instructor — online or in person. Free course lessons are for self-study and do not lead to a certificate — see registration options.