Training Models in Terrene
One of the core steps in creating a machine learning system with Terrene is training your model. With Terrene it is possible to train two different types of models General Purpose Predictive Models and Anomaly Detection Models. Predictive models are ideal for doing exactly as their name suggests, making predictions. Some of the common use cases for predictive models are energy forecasting, sales projections, inventory management, or predictive maintenance. Anomaly detection is used most commonly for detecting unexpected outcomes. For example, common uses for anomaly detection include detecting fraudulent transactions or monitoring for cyber attacks. Step by step instructions for creating either can be found by following the links below.
Once you have built your first model you may be interested in improving the model or changing certain aspects of it. More information can be found by following the links below.
If you are looking at training a model with our API information on training a model in python can be found here.