How to draw() the result of get_by_smooth? about gratia HOT 2 CLOSED

On Feb 15, 2019, at 11:45 AM, Gavin Simpson ***@***.***> wrote: I wouldn't use get_by_smooth() for this — it is exported only as a helper for developers wanting to identify and extract a by variable smoother. Use evaluate_smooth() or draw() on the fitted model and select the smooth you want. However, in trying draw() I am seeing a bug which I'll have to fix. For example: library('gratia') library('mgcv') set.seed(42) dat <- gamSim(4, n = 400, verbose = FALSE) m <- gam(y ~ fac + s(x2, by = fac) + s(x0), data = dat) draw(m, select = 's(x2)') The latter is supposed to work, but it isn't and I'll raise an issue to get this fixed. Following on from the above example you could use evaluate_smooth() to get back the data the draw() is extracting internally and then draw the plots yourself: es <- evaluate_smooth(m, smooth = 's(x2)') es library('dplyr') ## add 95% credible interval es <- mutate(es, lower = est - (2 * se), upper = est + (2 * se)) ## plot library('ggplot2') ggplot(es, aes(x = x2, y = est)) + geom_ribbon(aes(ymax = upper, ymin = lower, x = x2), inherit.aes = FALSE, alpha = 0.2) + geom_line() + facet_wrap(~ smooth) Which produces <https://user-images.githubusercontent.com/514696/52870895-81e3bd80-310e-11e9-85da-66babb415cf2.png> And then you can modify from there. Or do as draw() does and prepare separate plots for each smooth, but filtering the es object on the smooth column to select each of the smooths in turn and prepare a plot as above for each smooth separately and then use cowplot::plot_grid() to arrange them on the plot device. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#29 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AtcLoIP4JbvEezVs-hvL0ZH16M8HnhwKks5vNuQ3gaJpZM4a8pPN>.