yemale

Build prediction regions. Generate possible outcomes. Evaluate their consequences.

yemale builds multivariate conformal predictive distributions from predictions and observed outcomes. Use them to construct prediction regions, draw samples, assign probabilities to events, and compute expectations—for example, the expected cost of a decision.

The conformal construction provides finite-sample calibration without assuming a particular data distribution.

It uses optimal transport to handle scalar and vector-valued scores, without first reducing vectors to a scalar summary. The transport engine also works independently.

Install

python -m pip install yemale==0.1.0a2

Alpha release.

With your model

Start with a fitted model and held-out data X_cal, y_cal that were not used to train it. These data are used to calibrate the prediction regions.

from yemale import extend

model = extend(model)
model.conformalize(X_cal, y_cal)

cpd = model.predict_distribution(X_new)
region = cpd.region(0.9, rng=0)
region.contains(y_new)  # One Boolean per new outcome.

model.predict(X_new) still returns the original point predictions.

Already have predictions? Use yemale.conformalize(predictions, outcomes).

Sampling and summaries

Use the model's prediction as the fixed candidate for sampling:

candidate = model.predict(X_new[:1])[0]
cpd = model.predict_distribution(X_new[:1], candidate=candidate)
cpd.sample(1000, rng=0)
cpd.mean()
cpd.cov()

Explore

• Conformal prediction: use a model or prediction arrays, test outcomes, and select candidates.
• Optimal transport: fit point clouds, construct regions, and work with reference and predictive distributions.
• Mathematical guide: the construction and its connection to the API.
• Changelog.

License

Apache-2.0.