doc-min-dependencies is failing because the pad parameter of matplotlib.axes.Axes.set_title was introduced in matplotlib 2.2.0 whereas the min-dep env uses matplotlib 2.1.1

An alternative is just to use

ax1.text(s='Ridge regression \n with target transformation', x=-5e4, y=8e5, fontsize=12, multialignment='center')

instead. Though I note that the matplotlib recommended way to add a title to subplot is with matplotlib.axes.Axes.set_title.

This looks good. Personally I think it's more common to have y_pred on the x axis and y_true on the y axis for the scatter plot.

Could you please add a residual plot?

• y_pred - y_true on the y axis
• y_pred on the x axis.

I expect the residual plot without the TargetTransform to be "reverse-smile"/banana shaped which is a bad sign. With the target quantile transform, the banana should go away which means that the new model has a better fit.

However one should observe heteroschedastic noise on the residual plots (larger residuals absolute values for larger y_pred) which means that the least square loss modeling assumption are not meant. This hints that a better model would expect the variance of the residuals to increase with the expected mean value (y_pred). This could probably be better modeled via a Tweedie loss with p in range [1, 2].

Actually my second point on heteroschedastic noise is not that obvious with the Ames dataset. Maybe leave that analysis out. I would still love to see the residual plots :)

@ogrisel does this look okay?

whoops, wrong x axis!

@ogrisel

ping @ogrisel

Hi @lucyleeow rendering has some issues:

• the target distribution x-axis
• something went wrong in the last plot too
  Could you check? Thanks!

Thanks @cmarmo, I think i've fixed the plot problems!

The banana went away, that's cool :)

Thanks @lucyleeow