Handling Untrained Input in Wio Terminal Accelerometer-Based Machine Learning Model

I’m working on a machine learning model using accelerometer data from a Wio Terminal, trained on four specific classification labels. My concern is how to classify inputs that don’t match any of these trained labels.
Could anyone advise on the best approach to handle such untrained input? Specifically, I’m seeking strategies to ensure the model can effectively deal with data falling outside the trained categories.

1 Like