Feature Selection with Spectral Analysis

I have accelerometer data, a Spectral Analysis block, and a classification block. The spectral analysis give me a list of generated features and their importance. Is there a way to select which features to be used as input for the classifier?

Hi @jane,

On the “Create impulse” page, you can select which features are passed to a learning block (e.g. “Spectral features” under “Input features” on the “Classification” block in the image below).

However, you must use all of the output features from the processing block as your input–you cannot select individual features. You could create multiple processing blocks and select from there, but I know that’s not what you’re looking for.