Class OnnxInferenceNode

java.lang.Object
com.helix.core.parser.ast.OnnxInferenceNode
All Implemented Interfaces:
ExpressionNode

public class OnnxInferenceNode extends Object implements ExpressionNode
AST node representing an in-process machine learning inference invocation via ONNX Runtime.

Produces a scalar floating-point score (posterior probability or confidence score) within the range [0.0, 1.0] that can participate in downstream relational and boolean guard expressions (e.g., ML(fraud_model_v1) > 0.85).

  • Constructor Details

    • OnnxInferenceNode

      public OnnxInferenceNode(String modelName)
      Constructs an OnnxInferenceNode with default output tensor mapping ("probabilities", index 1).
      Parameters:
      modelName - unique name of the target ONNX model
    • OnnxInferenceNode

      public OnnxInferenceNode(String modelName, String outputTensorName, int outputIndex)
      Constructs an OnnxInferenceNode with custom output tensor mapping and index.
      Parameters:
      modelName - unique name of the target ONNX model
      outputTensorName - target output tensor identifier in the ONNX graph
      outputIndex - index within the output tensor to extract
  • Method Details

    • getModelName

      public String getModelName()
    • getOutputTensorName

      public String getOutputTensorName()
    • getOutputIndex

      public int getOutputIndex()
    • accept

      public <R> R accept(AstVisitor<R> visitor)
      Description copied from interface: ExpressionNode
      Accepts an AST visitor.
      Specified by:
      accept in interface ExpressionNode
      Type Parameters:
      R - return type
      Parameters:
      visitor - the visitor
      Returns:
      result from visitor
    • equals

      public boolean equals(Object o)
      Overrides:
      equals in class Object
    • hashCode

      public int hashCode()
      Overrides:
      hashCode in class Object
    • toString

      public String toString()
      Overrides:
      toString in class Object