All models

EEG representation · Core ML

CBraMod · Embeddings

Five seconds of EEG, as a vector.

The pretrained CBraMod backbone as a native Core ML embedder. A 14-channel, five-second window in; a 14,000-dimensional embedding out.

Open on Hugging Face
Input
[1, 14, 1000] at 200 Hz
Output
[1, 14000]
Precision
fp32
Parity
Worst rel-L2 1.3 × 10⁻⁵
License
BSD-3-Clause
Updated
19 Aug 2026

Overview

CBraMod is a foundation model for EEG that attends across channels and across time separately: criss-cross attention. This package is its pretrained backbone with the classification head removed, converted to Core ML.

A five-second, 14-channel window becomes a 14,000-dimensional embedding for any downstream decoder you train, on the device that records it.

Fig. 01 · Shapes

Fourteen channels, five patches, two hundred dimensions

Each one-second patch becomes a 200-dimensional token. Attention runs across channels and across time separately, then the tokens are flattened.

[1, 14, 1000]

14 channels × 5 s at 200 Hz

14 × 5 tokens

Criss-cross attention: across channels · across time

[1, 14000]

14 × 5 × 200 = 14,000

Embedding values are illustrative. The shapes are the package’s contract.

Validation

CheckResult
Embedding parity vs PyTorch, real EEGWorst relative L2 1.3 × 10⁻⁵
Motor-imagery decoding on Core ML embeddingsIdentical to PyTorch

Numerical parity is necessary, not sufficient, so the package was also checked at the task level: a motor-imagery decoder trained on its embeddings reaches the same decisions as one trained on PyTorch embeddings.

Input contract

Package
CBraModEmbedder.mlpackage
Input
eeg, float32 [1, 14, 1000]
Preprocessing
Average reference and global z-score at 256 Hz, then resample to 200 Hz
Output
embedding, float32 [1, 14000]

Know the boundaries

  • Fixed input shape: 14 channels and 1,000 samples. Channel count and preprocessing must match.
  • Embeddings are features. They are not calibrated intent probabilities, and a trained head is required for any decision.
  • Not clinically validated.

About this page

Summarized from the public model card and its published files. Repository updated 19 Aug 2026; reviewed 26 Sep 2026. The Hugging Face card is the source of truth for licenses and current details.

Upstream: CBraMod / braindecode