Class ONNXRuntimeBackend

java.lang.Object

org.episteme.core.mathematics.ml.neural.backends.ONNXRuntimeBackend

All Implemented Interfaces:

AlgorithmProvider, Backend

@AutoService({Backend.class,AlgorithmProvider.class})
public class ONNXRuntimeBackend
extends Object
implements Backend, AlgorithmProvider

Provider for running ONNX (Open Neural Network Exchange) models.

Allows executing pre-trained models (e.g., from PyTorch/TensorFlow) within Episteme. Implements Backend and AlgorithmProvider for standardized discovery.

Since:

2.0

Author:

Silvere Martin-Michiellot, Gemini AI (Google DeepMind)

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	ONNXRuntimeBackend.ONNXSession	Represents a loaded ONNX model session.

Constructor Summary

Constructors

Constructor	Description
ONNXRuntimeBackend()

Method Summary

Modifier and Type	Method	Description
Object	createBackend()	Creates and returns the backend instance (or returns self if it is the backend).
String	getAlgorithmType()	Returns the unique category of the algorithm.
String	getDescription()	Returns a description of the backend.
String	getId()	Returns the unique identifier for this backend (e.g., "javafx", "jmol", "cuda").
String	getName()	Returns the display name for UI presentation.
int	getPriority()	Returns the priority for auto-selection (higher = preferred).
String	getType()	Returns the backend type category (e.g., "plotting", "molecular", "tensor", "math").
boolean	isAvailable()	Checks if this backend is currently available (libraries loaded, etc.).
ONNXRuntimeBackend.ONNXSession	loadModel(Path modelPath)	Loads an ONNX model from the specified path.
double	score(OperationContext context)	Scores this provider for a specific operation context.
void	shutdown()	Called when the application is shutting down or the backend is being replaced.

Methods inherited from class Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface AlgorithmProvider

description, getMetadata

Methods inherited from interface Backend

getAlgorithmProviders, getStatusMessage, isExplicitlyDisabled

Constructor Details

ONNXRuntimeBackend

public ONNXRuntimeBackend()

Method Details

getType

public String getType()

Description copied from interface: Backend

Returns the backend type category (e.g., "plotting", "molecular", "tensor", "math").

Specified by:

getType in interface Backend

getId

public String getId()

Description copied from interface: Backend

Returns the unique identifier for this backend (e.g., "javafx", "jmol", "cuda").

Specified by:

getId in interface Backend

getName

public String getName()

Description copied from interface: Backend

Returns the display name for UI presentation.

Specified by:

getName in interface AlgorithmProvider

Specified by:

getName in interface Backend

getDescription

public String getDescription()

Description copied from interface: Backend

Returns a description of the backend.

Specified by:

getDescription in interface Backend

isAvailable

public boolean isAvailable()

Description copied from interface: Backend

Checks if this backend is currently available (libraries loaded, etc.).

Default implementation checks Backend.isExplicitlyDisabled() and returns false if it is.

Specified by:

isAvailable in interface AlgorithmProvider

Specified by:

isAvailable in interface Backend

getPriority

public int getPriority()

Description copied from interface: Backend

Returns the priority for auto-selection (higher = preferred). Used when multiple backends are available.

Specified by:

getPriority in interface AlgorithmProvider

Specified by:

getPriority in interface Backend

getAlgorithmType

public String getAlgorithmType()

Description copied from interface: AlgorithmProvider

Returns the unique category of the algorithm.

Specified by:

getAlgorithmType in interface AlgorithmProvider

score

public double score(OperationContext context)

Description copied from interface: AlgorithmProvider

Scores this provider for a specific operation context.

Higher scores indicate better suitability. Used by ProviderSelector for context-aware selection.

Default implementation returns AlgorithmProvider.getPriority(), so existing providers behave identically without changes.

Specified by:

score in interface AlgorithmProvider

Parameters:

context - the operation context (data size, hints, etc.)

Returns:

suitability score (higher = better)

createBackend

public Object createBackend()

Description copied from interface: Backend

Creates and returns the backend instance (or returns self if it is the backend).

Specified by:

createBackend in interface Backend

Returns:

The backend implementation object

loadModel

public ONNXRuntimeBackend.ONNXSession loadModel(Path modelPath)

Loads an ONNX model from the specified path.

Parameters:

modelPath - path to .onnx file.

Returns:

a runnable Model session.

shutdown

public void shutdown()

Description copied from interface: Backend

Called when the application is shutting down or the backend is being replaced. Use this to release hardware resources, close file handles, or stop background threads.

Specified by:

shutdown in interface AlgorithmProvider

Specified by:

shutdown in interface Backend