Is there a way to handle (ignore) y values that are None. If (for whatever reason) the computation that calculates the y value for a given x fails (produces None), is there a way to make a new guess for an x value (either based on previously captured data or with a random guess) and launch a new workflow?