PyTorch: Difference between revisions

From Tygron Support wiki
Jump to navigation Jump to search
No edit summary
Maxim@tygron.com (talk | contribs)
No edit summary
Line 3: Line 3:
PyTorch packages can be installed and managed using Conda. Conda-forge's Miniforge{{ref|Miniforge}} is an open-source application that allows you to create and manage conda environments.
PyTorch packages can be installed and managed using Conda. Conda-forge's Miniforge{{ref|Miniforge}} is an open-source application that allows you to create and manage conda environments.


Trained PyTorch Models can be exported using the [[ONNX]] file format, which in turn can be imported into the {{software}} as a [[Neural Network]] for an [[Inference Overlay]].
Trained PyTorch Models can be exported using the [[ONNX]] file format, which in turn can be imported into the {{software}} as a [[Region-based Convolutional Neural Network (Inference Overlay)|RCNN]] for an [[Inference Overlay]].




Line 17: Line 17:
* [[ONNX]]
* [[ONNX]]
* [[Inference Overlay]]
* [[Inference Overlay]]
* [[Neural Network]]
* [[Region-based Convolutional Neural Network (Inference Overlay)|RCNN]]
* [[Demo Training Data Project]]
* [[Demo Training Data Project]]



Revision as of 15:59, 29 June 2026

PyTorch[1] is a Python-based, open source, and production-ready AI framework that supports distributed training, graph mode, and cloud platforms.

PyTorch packages can be installed and managed using Conda. Conda-forge's Miniforge[2] is an open-source application that allows you to create and manage conda environments.

Trained PyTorch Models can be exported using the ONNX file format, which in turn can be imported into the Tygron Platform as a RCNN for an Inference Overlay.