VyntriVyntri
Solvers

Overview

Compare Vyntri adaptation and shrinkage options for fast, analytic vision intelligence.

Vyntri uses analytic ridge classification with optional feature adaptation — zero gradient descent, zero epochs, zero hyperparameter sweeps. Advanced adaptation and shrinkage options available for research use.

Quick Comparison

FK (Default)SLCELedoit-Wolf
Full NameFriedman's Discriminative WhiteningSupervised Linear Centroid-EncoderData-Estimated Shrinkage
How it worksEigenvalue flooring + diagonal shrinkageCentroid projection + eigenvalue flooringCross-validated shrinkage intensity
SpeedFastFastFast
Continual learning?YesNo (requires re-fit)No (requires re-fit)
Best forGeneral classificationResearch, centroid separationResearch, data-driven shrinkage

When To Use Which Method

ScenarioRecommendation
General ClassificationUse FK (default) — best overall performance with continual learning support.
Continual LearningUse FK + diagonal shrinkage — only supported combination for update().
Research / ExperimentsUse SLCE or Ledoit-Wolf — advanced options with honest trade-offs.
Auto BackboneUse backbone="auto" with optional candidates to control which backbones are evaluated.

Supported Backbones

All adaptation and shrinkage methods work identically across all five supported backbones: mobilenet_v3_small, resnet18, resnet50, efficientnet_b0, and convnext_tiny.