Embedding transforms

The Personalizer can apply an embedding-space transform on Z before the head. Use this when cal and stream embeddings shift, without retraining the trunk.

Families

Family	Names	When
Cal-fitted (unsupervised)	`standardize`, `whiten`, `euclidean_alignment`	Fit at `fit()` from calibration embeddings
Two-distribution / supervised	`coral`, `rpa`, `affine`	Need a stream (or second) slice after fit — offline/research, not deployment defaults

from nimbus_bci import Personalizer, wrap

enc = wrap(model.encode, model_id="partner", embedding_dim=64)

# Cal-fitted — common research path
p = Personalizer.for_research(enc, ["left", "right"], transform="whiten")
p.fit(X_cal, y_cal)

# Two-distribution maps — after fit, pass a shifted stream slice
p2 = Personalizer.for_research(enc, ["left", "right"], transform="coral")
p2.fit(X_cal, y_cal)
p2.fit_coral(X_stream)           # real two-distribution CORAL (Sun & Saenko 2016)
# p2.fit_rpa(X_stream, trim_windows=...) 
# p2.fit_affine(X_stream, y_stream)  # supervised L2 ridge toward identity

Or pass a stream at fit time when supported: fit(..., X_stream=...).

Quick guidance

Transform	Intent
`standardize`	Per-dim mean/std from cal
`whiten`	Decorrelate + unit variance from cal
`euclidean_alignment`	EA-style recentering in embedding space
`coral`	Align second-order stats to a target distribution
`rpa`	Re-centering with optional `trim_windows=` robustness
`affine`	Supervised affine map; ridge toward identity

EMA-damped label-free CORAL refits use ema_beta= on supported paths.

Prefer no transform (or cal-fitted only) for deployment. Treat coral / rpa / affine as research / offline alignment — they are not the product default.

Types

Public types: StandardizeTransform, WhitenTransform, CoralTransform, RpaTransform, EuclideanAlignment (also importable from nimbus_bci).

Next read

Paradigms & profiles

MI / P300 recipes and day-2 profiles

Encoder contract

wrap → Personalizer → save/load

Research / PEFT

BaLoRA and trunk-arm adaptation

API Reference

Full symbol list

Getting started

Personalizer

Python SDK

Inference & pipelines

Heads & models

Concepts

Julia SDK

Cloud API

Examples

Embedding transforms

Families

Quick guidance

Types

Next read

Paradigms & profiles

Encoder contract

Research / PEFT

API Reference

​Families

​Quick guidance

​Types

​Next read

Paradigms & profiles

Encoder contract

Research / PEFT

API Reference

Families

Quick guidance

Types

Next read