Appearance
Source files: 4 | Classes: 11 | Methods: 1 | Enums: 1
CognitiveCode.Silvia.Api
SilviaApiComponents
class
SILVIA Component System API - follows existing SILVIA API patterns
Provides component loading, execution, and management capabilities
Source: sv_apiComponents.cs
CognitiveCode.Silvia.Components
ComponentInfo
class
Complete metadata for a SILVIA component including parameters, inputs, and outputs.
This class contains all the information needed to instantiate, configure,
and connect a component. It is generated through reflection analysis
of component attributes and is used for UI generation and validation.
Performance: Metadata is cached after first analysis to avoid repeated reflection.
Source: SilviaComponentSystem.cs
InputInfo
class
Metadata for a data input on a SILVIA component.
Inputs represent data channels that must be connected to outputs
from other components. The component reads data from these inputs
during execution to perform its processing.
Source: SilviaComponentSystem.cs
ONNXInferenceComponent
class
ONNX Inference Component for SILVIA Framework.
This component provides high-performance AI/ML inference capabilities by loading
and executing ONNX (Open Neural Network Exchange) models. It supports various
execution modes, GPU acceleration, and flexible input/output tensor handling.
Key Features:
- ONNX Model Loading: Load and initialize ONNX models from file paths
- Flexible Execution: OnDemand, Continuous, Interval, and Triggered modes
- GPU Acceleration: Automatic GPU detection and CUDA provider setup
- Multi-threading: Configurable intra/inter-operator thread counts
- Tensor Conversion: Automatic conversion between .NET arrays and ONNX tensors
- Performance Optimized: Zero-allocation design with manual loop implementations
Usage Example:
Set ModelPath parameter to your .onnx file, configure execution mode,
provide input tensors via ModelInputs, and read results from ModelOutputs.
Thread Safety: Not thread-safe. Use one instance per execution context.
Source: ONNXInferenceComponent.cs
OutputInfo
class
Metadata for a data output on a SILVIA component.
Outputs represent data channels that will contain results after
component execution. Other components can connect to these outputs
to receive the processed data.
Source: SilviaComponentSystem.cs
ParameterInfo
class
Metadata for a configurable parameter on a SILVIA component.
Parameters are configuration values that can be set once and remain
constant during component execution. They are typically used for
component-specific settings and thresholds.
Source: SilviaComponentSystem.cs
SilviaComponent
abstract class
Base class for all SILVIA components
Source: SilviaComponentSystem.cs
CognitiveCode.Silvia.Unity
ComponentParameter
class
Serializable parameter for Unity Inspector
Source: SilviaUnityComponents.cs
Constants and Fields
boolValue
bool
colorValue
Color
description
string
floatValue
float
intValue
int
isAdvanced
bool
isRequired
bool
name
string
stringValue
string
type
ParameterType
vector3Value
Vector3
Methods
GetValue
object GetValue ( )
SilviaCoreWrapper
class
Unity MonoBehaviour wrapper for SILVIA Core instance
Source: SilviaUnityComponents.cs
SilviaONNXComponent
class
Unity MonoBehaviour wrapper for ONNX Inference Component
Source: SilviaUnityComponents.cs
Enumerations
ExecutionMode
Values: OnDemand, Continuous, Interval, Triggered
SilviaUnityComponentBase
abstract class
Base Unity MonoBehaviour wrapper for SILVIA components
Source: SilviaUnityComponents.cs
Generated from GTOS Savants source -- 2026-03-22
