Namespace MOSAIC.Models.Analytics
Classes
MetricsExtractor
Computes amplitude/error metrics for each channel in a windowed input signal.
OnlineICA
Online Independent Component Analysis (ICA) using incremental FastICA learning.
OnlineLDA
Online Linear Discriminant Analysis using incremental learning. Maximizes between-class scatter while minimizing within-class scatter.
Key difference from PCA:
- PCA finds directions of maximum variance (unsupervised)
- LDA finds directions of maximum class separation (supervised)
Inputs (like IncrementalPredictor):
- Input[0]: Data source block (provides feature vectors)
- Input[1]: Trigger block (provides class labels)
OnlineLDA.ClassStats
OnlinePCA
Online Principal Component Analysis using incremental (streaming) Oja's learning rule.
SupervisedICA
Supervised version of OnlineICA with manual cluster labelling and capture.
SupervisedPCA
Supervised version of OnlinePCA with manual cluster labelling and capture.
SupervisedUMAP
Base class for processing blocks in the MOSAIC pipeline.
Enums
MetricType
Available amplitude/error metric types.