The quality team read about deep learning and asks whether the risk flag should be rebuilt as a neural network. Give them a second opinion they can act on — backed by a notebook you wrote.
A starter notebook (project-starter.ipynb)
with load_and_recode(), split_by_patient(), and the MNIST loader already written, a
split_by_patient_3way(df, seed, val_size, test_size) stub (signature only — s4-2 taught
it, the body is yours), and one empty labelled cell per checkpoint.
Each checkpoint is one observable artifact you build yourself. Checking all six flips this module's mastery to "project complete" in the tracker. No auto-grading — you mark it once the artifact exists.
One held-out patient's Module 03 logistic-regression probability recomputed as a neuron in raw numpy and in PyTorch; three numbers agree to 4 decimals.
Uses the actual s2-2 weights, not refit; states in one sentence what changes when a second layer is added.
A 3-parameter chain (two weights + one bias) with both hand-derived gradients, PyTorch autograd's gradients, and a finite-difference check, all in one table.
The bias term derived (not copied from s3-1); the update rule quoted as Module 03's line; the finite-difference ε stated.
The MLP trained on a patient-safe three-way split with train and validation loss curves saved as a figure, and AUC + recall at matched flagged rate on the test rows.
Standardised on train statistics only; seed fixed; the validation slice never used for scoring; architecture drawn or listed.
An overfitting curve produced deliberately, then one countermeasure applied, before/after curves side by side, and one sentence on why it helps.
The stopping epoch/patience/p/λ chosen using validation, not test; the sentence names the mechanism (ensemble of thinned nets / capacity capped by time / weights taxed), not just "it regularises."
Four models (logistic, tree, MLP, boosted ensemble) on identical test rows, AUC and recall at matched flagged rate, each placed on the inspectability spectrum.
Reports the direction the data gives; places the ensemble and the MLP correctly on the spectrum; does not use accuracy (Module 03's 88.8% trap).
≤1 page, plain language, no code: should the flag move to a neural net? Answer first; evidence; what it would cost (data, compute, audit); the one situation where the answer flips.
Leads with the answer; cites the scoreboard numbers; names auditability as a cost in a hospital; the flip case is unstructured data (points at s5-1) and does not overclaim beyond it.
Widen the diabetes input with one-hot-encoded age and admission_type_id (categorical
encoding wasn't taught in this module — you're on your own here, which is the point) and see whether the
MLP's rank in the scoreboard moves.