It provides a more informative measure of variable importance than Garson’s algorithm. The ‘out_var’ argument also lets you choose which output node to evaluate.

For the activation function, I would look at the help files for the neural network method you’re using. Each package provides different options so you’ll have to take a look yourself.

]]>I suggest using the ‘olden’ function from the NeuralNetTools package: https://github.com/fawda123/NeuralNetTools

The Garson algorithm used here is outdated.

Best,

Marcus

]]>