search

autonlab / auton-survival

Auton Survival - an open source package for Regression, Counterfactual Estimation, Evaluation and Phenotyping with Censored Time-to-Events
http://autonlab.github.io/auton-survival
MIT License
315 stars 74 forks source link

Question about predictions on right-censored data #143

Open tyler-keller opened 1 month ago

tyler-keller commented 1 month ago

Great library, thanks so much for it!

I'm currently using the DeepRecurrentSurvivalMachines model and I'm trying to predict the time-to-event values for my right-censored data. From testing, the values it's returning are much smaller than I'd expect.

Does .predict_mean() return these values as I'm expecting?

  def predict_mean(self, x, risk=1):
    r"""Returns the mean Time-to-Event \( t \)

    Parameters
    ----------
    x: np.ndarray
        A numpy array of the input features, \( x \).
    Returns:
      np.array: numpy array of the mean time to event.
    """

If not, could you let me know how I could achieve this functionality?