Class HybridPredictorBlock
Hybrid Predictor Block - combines classifier training with continuous vector output.
Training: Like a classifier - learns from discrete class labels (action names from Trigger) Output: Continuous vector - weighted sum of class target vectors by predicted probabilities
Example:
Classes defined in Trigger:
rest: [0,0,0,0,0,0]
power: [1,1,1,1,0,0]
pinch: [0,0,0,0,1,1]
Predicted probabilities: [0.1, 0.7, 0.2]
Output = 0.1rest + 0.7power + 0.2*pinch = [0.7, 0.7, 0.7, 0.7, 0.2, 0.2]
This gives smooth, continuous control signals for prosthesis control while training with simple discrete gesture labels.
Inputs:
- Input[0]: Data source (feature vectors)
- Input[1]: Trigger (provides class labels AND target vectors)
JSON Config: { "Type": "HybridPredictor", "Inputs": ["DataSource", "Trigger"], "DesiredRate": 200, "Params": ["SoftmaxRFF", 8, 0.01, 0.001, 1.0, 300], "Path": "models/" }
Params: [modelType, inputDim, learningRate?, lambda?, sigma?, featureDim?] Note: Classes are automatically learned from Trigger actions!
JSON configuration
Use these block entries inside a pipeline configuration. Input names must match other blocks in that pipeline; device addresses and file paths must match your setup.
Block entry for a larger pipeline. Params: classifier model, feature dimension, learning rate, regularization, RFF kernel width, RFF feature count. Classes are discovered from Trigger action names; there is no numClasses parameter. Define an 8-value feature stream and a Trigger supplying targets.
{
"HybridPredictor": {
"Type": "hybridpredictor",
"Inputs": ["Features", "Capture"],
"Params": ["Softmax", 8, 0.01, 0.001, 1.0, 300]
}
}
How Type, Inputs and Params work · Block catalogue and parameter reference
Implements
Inherited Members
Namespace: MOSAIC.Models.Learning
Assembly: MOSAIC.dll
Syntax
public sealed class HybridPredictorBlock : BaseBlock, INotifyPropertyChanged, INotifyPropertyChanging, ISubscriber, IPublisher, IBaseBlockProps, IJsonExportable, IDisposable, IAsyncDisposable
Remarks
A BaseBlock represents one node in a processing graph. Blocks are typically created and linked based on a JSON configuration (pipeline definition).
Two Rate Properties:
| Property | Purpose |
|---|---|
| DesiredRate | Emission / display rate in Hz. Determines how often a block publishes output. Modified by windowing blocks (e.g. SlidingWindow: input rows/s divided by stride). Used for status monitoring and tick rate display. |
| SignalRate | True sample rate of the underlying signal in Hz. Used for DSP coefficient design (filter cutoffs, FFT bins, spectral resolution). Passes through windowing blocks unchanged — a SlidingWindow with stride 100 at 2000 Hz input still outputs data whose samples are at 2000 Hz, even though the windows arrive at 20 Hz. |
Rate Inheritance: Both rates propagate downstream automatically via TryInheritRate(BaseBlock). Publication rates initially fill unset values; use UpdateAndPropagateRate(double, bool) for runtime changes. Sample rates refresh downstream unless a block declares that it derives its own rate.
Publishing: The Publish(object) method is thread-safe and non-blocking. If a publish call is already in progress, the new value is dropped and the block enters a "stumbling" state reflected in Status.
Automatic CSV Logging: Every successful Publish(object) call feeds
Dumper whenever one is open. A dumper opens either because the config named a
Path — BlockFactory calls
InitDumper(IServiceProvider, string?) for every block it builds — or because IsRecording was
switched on, which writes to the shared session folder instead. Either way the file is named by
DumpFilePrefix. A block whose Path means something other than a recording
folder overrides PathIsDumpFolder.
Constructors
HybridPredictorBlock(string, double, int, ClassifierType, ClassifierConfig?, string?)
Declaration
public HybridPredictorBlock(string name, double desiredRate, int inputDim, ClassifierType modelType = ClassifierType.Softmax, ClassifierConfig? config = null, string? savePath = null)
Parameters
| Type | Name | Description |
|---|---|---|
| string | name | |
| double | desiredRate | |
| int | inputDim | |
| ClassifierType | modelType | |
| ClassifierConfig | config | |
| string | savePath |
Fields
_classIndices
Declaration
private readonly Dictionary<string, int> _classIndices
Field Value
| Type | Description |
|---|---|
| Dictionary<string, int> |
_classNames
Declaration
private readonly List<string> _classNames
Field Value
| Type | Description |
|---|---|
| List<string> |
_classProbabilities
Declaration
[ObservableProperty]
private double[] _classProbabilities
Field Value
| Type | Description |
|---|---|
| double[] |
_classVectors
Declaration
private readonly List<Vector<double>> _classVectors
Field Value
| Type | Description |
|---|---|
| List<Vector<double>> |
_confidence
Declaration
[ObservableProperty]
private double _confidence
Field Value
| Type | Description |
|---|---|
| double |
_currentClassName
Declaration
[ObservableProperty]
private string _currentClassName
Field Value
| Type | Description |
|---|---|
| string |
_currentLabel
Declaration
private int _currentLabel
Field Value
| Type | Description |
|---|---|
| int |
_isInitialized
Declaration
private bool _isInitialized
Field Value
| Type | Description |
|---|---|
| bool |
_isTraining
Declaration
[ObservableProperty]
[NotifyPropertyChangedFor("TrainingStatusText")]
private bool _isTraining
Field Value
| Type | Description |
|---|---|
| bool |
_lastOutput
Declaration
private Vector<double>? _lastOutput
Field Value
| Type | Description |
|---|---|
| Vector<double> |
_model
Declaration
private IClassifier _model
Field Value
| Type | Description |
|---|---|
| IClassifier |
_predictedClassName
Declaration
[ObservableProperty]
private string _predictedClassName
Field Value
| Type | Description |
|---|---|
| string |
_sampleCount
Declaration
[ObservableProperty]
private int _sampleCount
Field Value
| Type | Description |
|---|---|
| int |
_savePath
Declaration
private readonly string? _savePath
Field Value
| Type | Description |
|---|---|
| string |
_trainingMessage
Declaration
[ObservableProperty]
private string _trainingMessage
Field Value
| Type | Description |
|---|---|
| string |
Properties
ClassNames
Declaration
public IReadOnlyList<string> ClassNames { get; }
Property Value
| Type | Description |
|---|---|
| IReadOnlyList<string> |
ClassProbabilities
Declaration
[ExcludeFromCodeCoverage]
public double[] ClassProbabilities { get; set; }
Property Value
| Type | Description |
|---|---|
| double[] |
Confidence
Declaration
[ExcludeFromCodeCoverage]
public double Confidence { get; set; }
Property Value
| Type | Description |
|---|---|
| double |
Config
Declaration
public ClassifierConfig Config { get; private set; }
Property Value
| Type | Description |
|---|---|
| ClassifierConfig |
CurrentClassName
Declaration
[ExcludeFromCodeCoverage]
public string CurrentClassName { get; set; }
Property Value
| Type | Description |
|---|---|
| string |
InputDim
Declaration
public int InputDim { get; }
Property Value
| Type | Description |
|---|---|
| int |
IsTraining
Declaration
[ExcludeFromCodeCoverage]
public bool IsTraining { get; set; }
Property Value
| Type | Description |
|---|---|
| bool |
LastOutput
Gets the last output vector.
Declaration
public Vector<double>? LastOutput { get; }
Property Value
| Type | Description |
|---|---|
| Vector<double> |
MaxInputs
Gets the maximum number of input connections allowed for this block.
Declaration
public override int MaxInputs { get; }
Property Value
| Type | Description |
|---|---|
| int |
Overrides
Remarks
Override in derived classes to specify input requirements. Default is 1 (most blocks accept exactly one input). Use MaxValue for blocks that accept unlimited inputs (e.g., Joiner).
MinInputs
Gets the minimum number of input connections required for this block.
Declaration
public override int MinInputs { get; }
Property Value
| Type | Description |
|---|---|
| int |
Overrides
Remarks
Override in derived classes to specify input requirements. Default is 1 (most blocks require exactly one input). Source blocks (devices, generators) should return 0.
ModelType
Declaration
public ClassifierType ModelType { get; private set; }
Property Value
| Type | Description |
|---|---|
| ClassifierType |
NumClasses
Declaration
public int NumClasses { get; }
Property Value
| Type | Description |
|---|---|
| int |
OutputDim
Declaration
public int OutputDim { get; }
Property Value
| Type | Description |
|---|---|
| int |
PathIsDumpFolder
Path is this block's model directory, not a CSV destination.
Declaration
protected override bool PathIsDumpFolder { get; }
Property Value
| Type | Description |
|---|---|
| bool |
Overrides
Remarks
The only block in the pipeline that reads Path as something other than a recording
folder without also recording there. Without this the factory would drop a CSV in among the
saved model files.
PredictedClassName
Declaration
[ExcludeFromCodeCoverage]
public string PredictedClassName { get; set; }
Property Value
| Type | Description |
|---|---|
| string |
SampleCount
Declaration
[ExcludeFromCodeCoverage]
public int SampleCount { get; set; }
Property Value
| Type | Description |
|---|---|
| int |
TrainingMessage
Declaration
[ExcludeFromCodeCoverage]
public string TrainingMessage { get; set; }
Property Value
| Type | Description |
|---|---|
| string |
TrainingStatusText
Declaration
public string TrainingStatusText { get; }
Property Value
| Type | Description |
|---|---|
| string |
Methods
AddClass(string, Vector<double>)
Manually add a class with its target vector.
Declaration
public void AddClass(string name, Vector<double> targetVector)
Parameters
| Type | Name | Description |
|---|---|---|
| string | name | |
| Vector<double> | targetVector |
AutoSave()
Declaration
public void AutoSave()
ChangeModel(ClassifierType, ClassifierConfig?)
Changes the model type. Resets the model but keeps learned classes.
Declaration
public void ChangeModel(ClassifierType type, ClassifierConfig? config = null)
Parameters
| Type | Name | Description |
|---|---|---|
| ClassifierType | type | |
| ClassifierConfig | config |
ClearClasses()
Clears all learned classes and resets everything.
Declaration
public void ClearClasses()
ComputeWeightedOutput(double[])
Computes weighted sum of class vectors based on probabilities. Output = sum(prob[i] * classVector[i])
Declaration
private Vector<double> ComputeWeightedOutput(double[] probabilities)
Parameters
| Type | Name | Description |
|---|---|---|
| double[] | probabilities |
Returns
| Type | Description |
|---|---|
| Vector<double> |
ConfigureInput(IServiceProvider, JsonModel)
Creates block from JSON configuration.
Declaration
public static HybridPredictorBlock ConfigureInput(IServiceProvider sp, JsonModel m)
Parameters
| Type | Name | Description |
|---|---|---|
| IServiceProvider | sp | |
| JsonModel | m |
Returns
| Type | Description |
|---|---|
| HybridPredictorBlock |
Dispose()
Performs application-defined tasks associated with freeing, releasing, or resetting unmanaged resources.
Declaration
public override void Dispose()
Overrides
GetClassVector(int)
Gets the target vector for a class index.
Declaration
public Vector<double>? GetClassVector(int index)
Parameters
| Type | Name | Description |
|---|---|---|
| int | index |
Returns
| Type | Description |
|---|---|
| Vector<double> |
GetClassVector(string)
Gets the target vector for a class.
Declaration
public Vector<double>? GetClassVector(string name)
Parameters
| Type | Name | Description |
|---|---|---|
| string | name |
Returns
| Type | Description |
|---|---|
| Vector<double> |
GetDefaultSavePath()
Declaration
private string? GetDefaultSavePath()
Returns
| Type | Description |
|---|---|
| string |
GetDouble(object?, double)
Declaration
private static double GetDouble(object? obj, double defaultValue = 0)
Parameters
| Type | Name | Description |
|---|---|---|
| object | obj | |
| double | defaultValue |
Returns
| Type | Description |
|---|---|
| double |
GetInt(object?, int)
Declaration
private static int GetInt(object? obj, int defaultValue = 0)
Parameters
| Type | Name | Description |
|---|---|---|
| object | obj | |
| int | defaultValue |
Returns
| Type | Description |
|---|---|
| int |
GetJsonParams()
The six values ConfigureInput reads, in its order. Taken from the live model and config
rather than the constructor arguments, so a model swapped at run time is what gets saved.
Declaration
protected override IReadOnlyList<object>? GetJsonParams()
Returns
| Type | Description |
|---|---|
| IReadOnlyList<object> |
Overrides
GetJsonPath()
Puts the model directory back in the config on save.
Declaration
protected override string? GetJsonPath()
Returns
| Type | Description |
|---|---|
| string |
Overrides
Remarks
ToJsonModel() exports GetJsonPath() ?? _explicitDumpFolder, and
this block opens no dumper, so without the override both halves were null and saving a
pipeline silently dropped the model directory.
GetString(object?, string)
Declaration
private static string GetString(object? obj, string defaultValue = "")
Parameters
| Type | Name | Description |
|---|---|---|
| object | obj | |
| string | defaultValue |
Returns
| Type | Description |
|---|---|
| string |
HandleDataInput(Vector<double>)
Declaration
private void HandleDataInput(Vector<double> x)
Parameters
| Type | Name | Description |
|---|---|---|
| Vector<double> | x |
HandleTriggerInput(object, object?)
Declaration
private void HandleTriggerInput(object sender, object? data)
Parameters
| Type | Name | Description |
|---|---|---|
| object | sender | |
| object | data |
LearnNewClass(string, Vector<double>)
Declaration
private void LearnNewClass(string name, Vector<double> targetVector)
Parameters
| Type | Name | Description |
|---|---|---|
| string | name | |
| Vector<double> | targetVector |
LoadModel(string)
Declaration
public void LoadModel(string path)
Parameters
| Type | Name | Description |
|---|---|---|
| string | path |
OnClassProbabilitiesChanged(double[])
Executes the logic for when ClassProbabilities just changed.
Declaration
private void OnClassProbabilitiesChanged(double[] value)
Parameters
| Type | Name | Description |
|---|---|---|
| double[] | value | The new property value that was set. |
Remarks
This method is invoked right after the value of ClassProbabilities is changed.
OnClassProbabilitiesChanged(double[]?, double[])
Executes the logic for when ClassProbabilities just changed.
Declaration
private void OnClassProbabilitiesChanged(double[]? oldValue, double[] newValue)
Parameters
| Type | Name | Description |
|---|---|---|
| double[] | oldValue | The previous property value that was replaced. |
| double[] | newValue | The new property value that was set. |
Remarks
This method is invoked right after the value of ClassProbabilities is changed.
OnClassProbabilitiesChanging(double[])
Executes the logic for when ClassProbabilities is changing.
Declaration
private void OnClassProbabilitiesChanging(double[] value)
Parameters
| Type | Name | Description |
|---|---|---|
| double[] | value | The new property value being set. |
Remarks
This method is invoked right before the value of ClassProbabilities is changed.
OnClassProbabilitiesChanging(double[]?, double[])
Executes the logic for when ClassProbabilities is changing.
Declaration
private void OnClassProbabilitiesChanging(double[]? oldValue, double[] newValue)
Parameters
| Type | Name | Description |
|---|---|---|
| double[] | oldValue | The previous property value that is being replaced. |
| double[] | newValue | The new property value being set. |
Remarks
This method is invoked right before the value of ClassProbabilities is changed.
OnConfidenceChanged(double)
Executes the logic for when Confidence just changed.
Declaration
private void OnConfidenceChanged(double value)
Parameters
| Type | Name | Description |
|---|---|---|
| double | value | The new property value that was set. |
Remarks
This method is invoked right after the value of Confidence is changed.
OnConfidenceChanged(double, double)
Executes the logic for when Confidence just changed.
Declaration
private void OnConfidenceChanged(double oldValue, double newValue)
Parameters
| Type | Name | Description |
|---|---|---|
| double | oldValue | The previous property value that was replaced. |
| double | newValue | The new property value that was set. |
Remarks
This method is invoked right after the value of Confidence is changed.
OnConfidenceChanging(double)
Executes the logic for when Confidence is changing.
Declaration
private void OnConfidenceChanging(double value)
Parameters
| Type | Name | Description |
|---|---|---|
| double | value | The new property value being set. |
Remarks
This method is invoked right before the value of Confidence is changed.
OnConfidenceChanging(double, double)
Executes the logic for when Confidence is changing.
Declaration
private void OnConfidenceChanging(double oldValue, double newValue)
Parameters
| Type | Name | Description |
|---|---|---|
| double | oldValue | The previous property value that is being replaced. |
| double | newValue | The new property value being set. |
Remarks
This method is invoked right before the value of Confidence is changed.
OnCurrentClassNameChanged(string)
Executes the logic for when CurrentClassName just changed.
Declaration
private void OnCurrentClassNameChanged(string value)
Parameters
| Type | Name | Description |
|---|---|---|
| string | value | The new property value that was set. |
Remarks
This method is invoked right after the value of CurrentClassName is changed.
OnCurrentClassNameChanged(string?, string)
Executes the logic for when CurrentClassName just changed.
Declaration
private void OnCurrentClassNameChanged(string? oldValue, string newValue)
Parameters
| Type | Name | Description |
|---|---|---|
| string | oldValue | The previous property value that was replaced. |
| string | newValue | The new property value that was set. |
Remarks
This method is invoked right after the value of CurrentClassName is changed.
OnCurrentClassNameChanging(string)
Executes the logic for when CurrentClassName is changing.
Declaration
private void OnCurrentClassNameChanging(string value)
Parameters
| Type | Name | Description |
|---|---|---|
| string | value | The new property value being set. |
Remarks
This method is invoked right before the value of CurrentClassName is changed.
OnCurrentClassNameChanging(string?, string)
Executes the logic for when CurrentClassName is changing.
Declaration
private void OnCurrentClassNameChanging(string? oldValue, string newValue)
Parameters
| Type | Name | Description |
|---|---|---|
| string | oldValue | The previous property value that is being replaced. |
| string | newValue | The new property value being set. |
Remarks
This method is invoked right before the value of CurrentClassName is changed.
OnIsTrainingChanged(bool)
Executes the logic for when IsTraining just changed.
Declaration
private void OnIsTrainingChanged(bool value)
Parameters
| Type | Name | Description |
|---|---|---|
| bool | value | The new property value that was set. |
Remarks
This method is invoked right after the value of IsTraining is changed.
OnIsTrainingChanged(bool, bool)
Executes the logic for when IsTraining just changed.
Declaration
private void OnIsTrainingChanged(bool oldValue, bool newValue)
Parameters
| Type | Name | Description |
|---|---|---|
| bool | oldValue | The previous property value that was replaced. |
| bool | newValue | The new property value that was set. |
Remarks
This method is invoked right after the value of IsTraining is changed.
OnIsTrainingChanging(bool)
Executes the logic for when IsTraining is changing.
Declaration
private void OnIsTrainingChanging(bool value)
Parameters
| Type | Name | Description |
|---|---|---|
| bool | value | The new property value being set. |
Remarks
This method is invoked right before the value of IsTraining is changed.
OnIsTrainingChanging(bool, bool)
Executes the logic for when IsTraining is changing.
Declaration
private void OnIsTrainingChanging(bool oldValue, bool newValue)
Parameters
| Type | Name | Description |
|---|---|---|
| bool | oldValue | The previous property value that is being replaced. |
| bool | newValue | The new property value being set. |
Remarks
This method is invoked right before the value of IsTraining is changed.
OnPredictedClassNameChanged(string)
Executes the logic for when PredictedClassName just changed.
Declaration
private void OnPredictedClassNameChanged(string value)
Parameters
| Type | Name | Description |
|---|---|---|
| string | value | The new property value that was set. |
Remarks
This method is invoked right after the value of PredictedClassName is changed.
OnPredictedClassNameChanged(string?, string)
Executes the logic for when PredictedClassName just changed.
Declaration
private void OnPredictedClassNameChanged(string? oldValue, string newValue)
Parameters
| Type | Name | Description |
|---|---|---|
| string | oldValue | The previous property value that was replaced. |
| string | newValue | The new property value that was set. |
Remarks
This method is invoked right after the value of PredictedClassName is changed.
OnPredictedClassNameChanging(string)
Executes the logic for when PredictedClassName is changing.
Declaration
private void OnPredictedClassNameChanging(string value)
Parameters
| Type | Name | Description |
|---|---|---|
| string | value | The new property value being set. |
Remarks
This method is invoked right before the value of PredictedClassName is changed.
OnPredictedClassNameChanging(string?, string)
Executes the logic for when PredictedClassName is changing.
Declaration
private void OnPredictedClassNameChanging(string? oldValue, string newValue)
Parameters
| Type | Name | Description |
|---|---|---|
| string | oldValue | The previous property value that is being replaced. |
| string | newValue | The new property value being set. |
Remarks
This method is invoked right before the value of PredictedClassName is changed.
OnReceive(object, object)
Handles incoming data from connected blocks.
Declaration
protected override void OnReceive(object sender, object data)
Parameters
| Type | Name | Description |
|---|---|---|
| object | sender | |
| object | data |
Overrides
OnSampleCountChanged(int)
Executes the logic for when SampleCount just changed.
Declaration
private void OnSampleCountChanged(int value)
Parameters
| Type | Name | Description |
|---|---|---|
| int | value | The new property value that was set. |
Remarks
This method is invoked right after the value of SampleCount is changed.
OnSampleCountChanged(int, int)
Executes the logic for when SampleCount just changed.
Declaration
private void OnSampleCountChanged(int oldValue, int newValue)
Parameters
| Type | Name | Description |
|---|---|---|
| int | oldValue | The previous property value that was replaced. |
| int | newValue | The new property value that was set. |
Remarks
This method is invoked right after the value of SampleCount is changed.
OnSampleCountChanging(int)
Executes the logic for when SampleCount is changing.
Declaration
private void OnSampleCountChanging(int value)
Parameters
| Type | Name | Description |
|---|---|---|
| int | value | The new property value being set. |
Remarks
This method is invoked right before the value of SampleCount is changed.
OnSampleCountChanging(int, int)
Executes the logic for when SampleCount is changing.
Declaration
private void OnSampleCountChanging(int oldValue, int newValue)
Parameters
| Type | Name | Description |
|---|---|---|
| int | oldValue | The previous property value that is being replaced. |
| int | newValue | The new property value being set. |
Remarks
This method is invoked right before the value of SampleCount is changed.
OnTrainingMessageChanged(string)
Executes the logic for when TrainingMessage just changed.
Declaration
private void OnTrainingMessageChanged(string value)
Parameters
| Type | Name | Description |
|---|---|---|
| string | value | The new property value that was set. |
Remarks
This method is invoked right after the value of TrainingMessage is changed.
OnTrainingMessageChanged(string?, string)
Executes the logic for when TrainingMessage just changed.
Declaration
private void OnTrainingMessageChanged(string? oldValue, string newValue)
Parameters
| Type | Name | Description |
|---|---|---|
| string | oldValue | The previous property value that was replaced. |
| string | newValue | The new property value that was set. |
Remarks
This method is invoked right after the value of TrainingMessage is changed.
OnTrainingMessageChanging(string)
Executes the logic for when TrainingMessage is changing.
Declaration
private void OnTrainingMessageChanging(string value)
Parameters
| Type | Name | Description |
|---|---|---|
| string | value | The new property value being set. |
Remarks
This method is invoked right before the value of TrainingMessage is changed.
OnTrainingMessageChanging(string?, string)
Executes the logic for when TrainingMessage is changing.
Declaration
private void OnTrainingMessageChanging(string? oldValue, string newValue)
Parameters
| Type | Name | Description |
|---|---|---|
| string | oldValue | The previous property value that is being replaced. |
| string | newValue | The new property value being set. |
Remarks
This method is invoked right before the value of TrainingMessage is changed.
ReinitializeClassifier()
Declaration
private void ReinitializeClassifier()
ResetModel()
Resets the model but keeps learned classes.
Declaration
public void ResetModel()
SaveModel(string?)
Declaration
public void SaveModel(string? path = null)
Parameters
| Type | Name | Description |
|---|---|---|
| string | path |
Events
OnClassLearned
Event fired when a new class is learned.
Declaration
public event Action<string, Vector<double>>? OnClassLearned
Event Type
| Type | Description |
|---|---|
| Action<string, Vector<double>> |
OnPrediction
Event fired when a new prediction is made. Args: (outputVector, predictedClass, probabilities, confidence, sampleCount)
Declaration
public event Action<Vector<double>, int, double[], double, int>? OnPrediction
Event Type
| Type | Description |
|---|---|
| Action<Vector<double>, int, double[], double, int> |
OnTrainingStateChanged
Event fired when training state changes.
Declaration
public event Action<bool, string?>? OnTrainingStateChanged
Event Type
| Type | Description |
|---|---|
| Action<bool, string> |