Is it easy to support variable weight using point size?

[ShixiangWang]

In the model, weight may contain important information. However, in current forestmodel, the point size is fixed.
So I wonder it would be nice to implement this.

forestmodel/R/panel_forest_plot.R

Lines 350 to 354 in dd8eb55

	main_plot <- main_plot +
	geom_point(aes(x, y),
	size = format_options$point_size, filter(mapped_data, !diamond),
	colour = format_options$colour, shape = format_options$shape, na.rm = TRUE
	)

[NikNakk]

How would you propose scaling the points? I understand for a meta-analysis forest plot, but not one for a glm.

[ShixiangWang]

@NikNakk Yeah, this is true. Some models may find it useful, but some others may not.

[NikNakk]

Happy to review suggested code (or even just a suggested algorithm for calculating point size).

[ShixiangWang]

Thanks, I will look into it in my spare time.