ONNX: Difference between revisions

From Tygron Support wiki
Jump to navigation Jump to search
No edit summary
Maxim@tygron.com (talk | contribs)
No edit summary
 
(5 intermediate revisions by the same user not shown)
Line 1: Line 1:
The Open [[Neural Network]] Exchange (ONNX){{ref|ONNX}} is an open-source artificial intelligence format for among other things representing machine learning algorithms.
The Open Neural Network Exchange (ONNX){{ref|ONNX}} is an open-source artificial intelligence format for among other things representing machine learning algorithms.
[[PyTorch]] models can be exported to ONNX files and used as an input for the [[Inference Overlay]].
[[PyTorch]] models can be exported to ONNX files and used as an input for the [[Inference Overlay]].


Dragging an ONNX file onto the Tygron Client Application allows you to:
Dragging an ONNX file onto the Tygron Client Application allows you to:
* Import it as a [[Neural Network]] in your project
* Import it as a [[Region-based Convolutional Neural Network (Inference Overlay)|Region-based Convolutional Neural Network (RCNN)]] in your project
* Import it with a new [[Inference Overlay]]. This also imports it as a new Neural Network and configures it automatically for the newly added Inference Overlay
* Import it with a new [[Inference Overlay]]. This also imports it as a new Neural Network and configures it automatically for the newly added Inference Overlay


Line 9: Line 9:
|seealso=
|seealso=
* [[Inference Overlay]]  
* [[Inference Overlay]]  
* [[Neural Network]]
* [[Region-based Convolutional Neural Network (Inference Overlay)|RCNN]]
* [[PyTorch]]
* [[PyTorch]]
* [[Demo Training Data Project]]
* [[Demo Training Data Project]]

Latest revision as of 16:06, 29 June 2026

The Open Neural Network Exchange (ONNX)[1] is an open-source artificial intelligence format for among other things representing machine learning algorithms. PyTorch models can be exported to ONNX files and used as an input for the Inference Overlay.

Dragging an ONNX file onto the Tygron Client Application allows you to: