Home
Portfolio
Professor Review
Download Deck
ReneWind · Introduction to Neural Networks

Professor Review

A rubric-aligned executive summary of the wind turbine generator failure-prediction project, with evidence links to every page that produced each result.

Download submission deck
Enter the access code to download the full presentation.

Headline numbers

Test recall
87.59%
247 of 282 failures caught
Test accuracy
98.74%
only 28 false alarms
Test F1 · AUC
88.69% · 0.939
precision 89.82%
Cost avoided
$10.98M
64.9% below a no-model baseline

Validation recall and F1 across all ten configurations

M8 — 128-64-32, dropout 0.3, Adam, class weights — leads on recall at 91.89% while holding precision above 91%.

Rubric coverage

Reproducibility

Model assumptions and settings
  • Train / validation split: 80 / 20, stratified on Target, random_state 42
  • Imputation: median, fitted on the training split only
  • Scaling: StandardScaler, fitted on the training split only
  • Hidden activation / init: ReLU with He-normal initialisation
  • Output / loss: sigmoid with binary cross-entropy
  • Epochs / batch size: 30 / 128 for every configuration
  • Class weights: failure 9.009, no failure 0.529
  • Decision threshold: 0.5 for all reported metrics
  • Random seed: 42 (numpy, TensorFlow and Keras)

Evaluator Review

Evaluator review workflow
Enter the review access code to capture evaluator notes and store them as a numbered version alongside this submission.