Hi Gavin,

Could you say more about the `partial_residuals()` function? Is this providing the residuals after accounting for the other covariates in the model or the residuals accounting for the smooth term? I am wondering because I would like to be able to plot the fitted smooth against the response variable that is adjusted for the other covariates, as in an adjusted response plot. After reading this post, I am trying to understand if there is a handy way to combine `partial_residuals()` with `smooth_estimates()` to do this. Would love to hear your thoughts.

Of course, thank you so much for developing the gratia package. It is awesome.

I'm looking forward to trying out gratia as soon as. Thank you Gavin.
I wonder though, whether the draw.gam() function, if a constant is passed in, should switch the y axis label from "effect" to "outcome" or something else. Or do you think it's explicit enough that it is no longer a partial and additive effect?