ISAmigrateModel

Update a model from an earlier toolbox version to the current layout

Syntax

model = ISAmigrateModel(rootdir)

model = ISAmigrateModel(rootdir,'backupSuffix',suffix)

model = ISAmigrateModel(model)

Description

model = ISAmigrateModel(rootdir) migrates rootdir/model.mat in place. The original file is first copied to model_legacy.mat in the same folder. Use this form once, to convert a saved model.

model = ISAmigrateModel(rootdir,'backupSuffix',suffix) names the backup model<suffix>.mat. An existing backup is never overwritten; ISAmigrateModel raises ISA:ISAmigrateModel:backupExists instead.

model = ISAmigrateModel(model) migrates a model structure already in memory and writes no files. InstanceSpace.load and exploreIS use this form on every load, so a legacy model.mat also works without converting it first.

Examples

Convert a saved model

ISAmigrateModel('/path/to/old/run/');
obj = InstanceSpace.load('/path/to/old/run/');

Migrate in memory

m = load('/path/to/old/run/model.mat');
m = ISAmigrateModel(m);
m.pythia.classifiers

Input Arguments

rootdir — Folder containing model.mat

character vector | string scalar

model — Model structure

structure

A model loaded with load('model.mat').

suffix — Backup file suffix

'_legacy' (default) | character vector

File form only.

Output Arguments

model — Migrated model

structure

Algorithms

Legacy content	Becomes
`opts.oracle`, `opts.pbldr`, `opts.sbound`, `opts.footprint`	`opts.pythia`, `opts.pilot`, `opts.cloister`, `opts.trace`
`opts.corr.flag`, `opts.corr.threshold`, `opts.clust.flag`	fields of `opts.sifted`
`opts.perf.MaxMin`	`opts.perf.MaxPerf`
`model.data.bestPerformace`	`model.data.Ybest`
PLS model without `pilot.Xmean`	recover the mean from training features and translate saved `cloist.Zedge`/`Zecorr` by `-Xmean*A'`; face indices are unchanged
`model.pythia.svm{i}`, `model.pythia.knn{i}`	`model.pythia.classifiers{i}`
`model.pythia.boxcosnt`, `model.pythia.kscale`	`model.pythia.param1`, `model.pythia.param2`
LIBSVM classifier structures	retrained with `opts.pythia.classifier` (default `'knn'`), because LIBSVM structures have no `predict` method
`model.trace` in the DBSCAN/polyshape format	recomputed with TRACE3, using `model.pythia.Yhat` when available, otherwise `model.data.Ybin`
no `completedStages`	inferred from the stage fields present

A model.pilot.A without B and C cannot be repaired and raises a warning.

Version History

Unreleased — PLS boundary alignment

Recovering a missing PLS feature mean also translates any saved CLOISTER vertices into the fitted PLS coordinate system. Empty or absent boundaries are retained. Repeated migration leaves already centred geometry unchanged. Models without the training features needed to recover the mean must be rebuilt.