Lantern Court

Compare models with clarity.

Side-by-side metrics, thoughtful caveats, and statistical context for honest decisions.

Live on

eight real models, ranked live in your browser

Kernel asleep

ROC AUCAccuracyF1

Churn v1 (Test) ⌄

Filters

Export

#	Model	Accuracy	Acc	F1	Time
Booting the kernel and ranking eight real models…

Eight scikit-learn classifiers · 5-fold cross-validationComputed in your browser via Pyodide

Rank #2

Strong overall performance with good balance across metrics and efficiency.

ROC AUC

0.912

Accuracy

0.876

F1

0.875

Time (s)

0.74

Strengths

Best ROC AUC, significantly better than the majority.

Caveats

Accuracy not significantly better than #1.

Reproduce

from breezeml import Breeze
data = Breeze.load("churn_v1")
model = Breeze.setup(model="xgboost")
model.train(data)
model.evaluate()

Next: cross the Honest Gate. A high score means nothing until it survives scrutiny.