AI Suite
Machine Learning
Using the Inference Overlay (available in version 2025.0+) users can do machine learning on public and private data.
For example detecting trees, bushes or objects from satellite imagery. Users can select an existing neural network and apply it directly on their own projects.
The neural network will be executed directly on our own Tygron Supercomputer GPU’s.
Open Neural Network Exchange
The Tygron Platform supports the Open Neural Network Exchange (ONNX) format. This allow users to also upload their own trained models, trained in for example PyTorch and use them to run inference directly in the Tygron Engine.
AI Training Data Creator
The Tygron Platform can also help with creating your own Training Data for PyTorch.
First select objects that must be recognized form e.g. trees in satellite imagery via QGIS connected with the Tygron Engine.
Then export the images, labels and mask directly from your projects as Training Data. Images can be satellite imagery, but also a many different types, like flow patterns or height maps.
Python AI SDK
We also have a public open-source repository on GitHub then can help you with the creation and training of the correct Neural Networks.
Following these steps allows you will train your own neural network that can then be uploaded into the Tygron Engine for inference.
Tygron and Artificial Intelligence
by Maxim Knepfle, CTO Tygron Artificial Intelligence (AI) is hot, and it seems that everyone needs to implement some sort of AI. But what does AI entail, what is the
