MODULE 05

Data audits and training diagnosis

Repair an evaluation pipeline and diagnose a training failure, with separate project assessments.

What you’ll explore

Original Learn AI lessons with local project assessments.

0 / 10 complete

Define inputs, outputs, users and success before choosing a model. Test how a changed target changes the labels.

Inspect missing values, copied requests, unit mismatches and conflicting labels before treating a table as evidence.

Compare a rule, supplied candidate predictions and simulated human review, including costs and unfinished work.

Audit real flower measurements, separate related rows, and turn a tempting score into a defensible claim.

Inspect predictions, gradients and actual parameter changes. Compare a failure and repair from the same checkpoint.

Move a threshold, count different errors, and defend a decision using precision, recall and explicit costs.

Draw repeated samples, inspect uncertainty and discover why more biased data can strengthen the wrong claim.

Trace a limited review budget through two channels and compare who is missed, reviewed unnecessarily or absent from the evidence.

Choose justified fields, examine provenance and permission, and build a practice data card with explicit handling limits.

Turn results into bounded claims, identify missing evidence and draft a model card linked to your actual experiment.