Objeck Documentation

Bundle

ONNX Runtime inference library for running pre-trained models. Supports object detection (YOLO), image classification (ResNet), segmentation (DeepLab), pose estimation (OpenPose), face recognition, and Phi-3 text generation. Compile with -lib onnx.

RunResult

The tensors a run produced. `Session->Run` returns these as a Vector and most callers want that. This is the same data before it is split apart, for a caller that would rather read one output out of many without building objects for the rest.

Get #

One output by name.

method : public : Get(name:String) ~ Tensor

Parameters

Name	Type	Description
name	String	the output name

Return

Type	Description
Tensor	the tensor, or Nil when the run produced no such output

GetCount #

How many tensors the run produced.

method : public : GetCount() ~ Int

Return

Type	Description
Int	output count

GetNames #

Output names, in the order they were produced.

method : public : GetNames() ~ String[]

Return

Type	Description
String	names

GetTensors #

Every output as a tensor.

method : public : GetTensors() ~ Vector<Tensor>

Return

Type	Description
Vector<Tensor>	output tensors