varImportance

This function calculates relative variable importance by drop-out. The values of a variable are set to its mean and the log likelihood is recalculated. The difference between the original log likelihood and the new log likelihood is the variable importance.

Parameters:

• model - the model returned by trainCoxMlp
• x_train - training set matrix. Expected numpy array.
• ytime_train - time of death or censoring for each patient
• ystatus_train - censoring of each patient