PREDICT. EXPERIMENT. REFLECT.
Learn to question
the model,
one experiment
at a time.
Change the reference window, shift the data, or adjust the review threshold. See which labels move—and which records the model cannot score.
GUIDED LEARNING / THREE CHECKPOINTS
Before you touch the slider,
make a prediction.
These lessons always use the built-in synthetic example. The free explorer below is separate. Journal entries stay in this tab until exported; no student account or learning analytics is collected.
Informal self-assessment only. No certificate, graded outcome or measured learning improvement is claimed.
01 / CHANGE AN ASSUMPTION
The controls
A higher threshold generally flags fewer records. It does not prove the model is better.
Uses the first N examples in each route, preserving input order. Fewer than eight cannot support a route score.
Adds minutes to reference transit values only. Candidates stay fixed. This is a controlled distribution-shift experiment.
02 / INSPECT THE CONSEQUENCE
Same records.
Different assumptions.
| Record | Original | Experiment | Distance / threshold |
|---|
Click a record to inspect its nearest reference examples. “Within reference” means within a fitted distance range, not safe, normal or clinically acceptable.
NEIGHBOR EVIDENCE
Select a record
03 / KEEP THE LIMITS VISIBLE
Experiment, then ask better questions.
What the model measures
Three-nearest-neighbor mean distance over transit minutes, queue minutes and missing scans. Features use robust MAD scaling with a floor. The original threshold is the leave-one-out reference 95th percentile. Every route needs at least eight reference examples.
What this cannot tell you
There are no truth labels, so the lab cannot calculate accuracy, precision, recall or patient outcomes. The selected window is order-dependent and may be biased. No automated decision should follow these teaching labels.
New work and prior work
The kNN model comes from Amrit Lahari's Custody Lens, adapted from MED-BLACKBOX. The explorer and experiment engine were built previously for Hyperbloom. New ML Empowerment work is the guided prediction workflow, model-backed feedback, preserved incorrect answers, optional reflections and exportable learning journal. OpenAI Codex assisted with code, tests and documentation.
Imports are processed in the current tab and never saved by this app. Use synthetic or non-sensitive data only; IDs and routes are not anonymized. Maximum 200 references, 100 candidates, 2 MB.