{"id":3442,"date":"2015-05-26T17:04:44","date_gmt":"2015-05-26T22:04:44","guid":{"rendered":"http:\/\/rud.is\/b\/?p=3442"},"modified":"2018-03-07T16:43:48","modified_gmt":"2018-03-07T21:43:48","slug":"a-quick-incomplete-comparison-of-ggplot2-rbokeh-plotting-idioms","status":"publish","type":"post","link":"https:\/\/rud.is\/b\/2015\/05\/26\/a-quick-incomplete-comparison-of-ggplot2-rbokeh-plotting-idioms\/","title":{"rendered":"A quick, incomplete comparison of ggplot2 & rbokeh plotting idioms"},"content":{"rendered":"

I set aside a small bit of time to give [rbokeh](https:\/\/github.com\/bokeh\/rbokeh) a try and figured I’d share a small bit of code that shows how to make the “same” chart in both ggplot2 and rbokeh. <\/p>\n

#### What is Bokeh\/rbokeh?<\/p>\n

rbokeh is an [htmlwidget](http:\/\/htmlwidgets.org) wrapper for the [Bokeh](http:\/\/bokeh.pydata.org\/en\/latest\/) visualization library that has become quite popular in Python circles. Bokeh makes creating interactive charts pretty simple and rbokeh lets you do it all with R syntax.<\/p>\n

#### Comparing ggplot & rbokeh<\/p>\n

This is not a comprehensive introduction into rbokeh. You can get that [here (officially)](http:\/\/hafen.github.io\/rbokeh\/). I merely wanted to show how a ggplot idiom would map to an rbokeh one for those that may be looking to try out the rbokeh library and are familiar with ggplot. They share a very common “grammar of graphics” base where you have a plot structure and add layers and aesthetics. They each do this a tad bit differently, though, as you’ll see.<\/p>\n

First, let’s plot a line graph with some markers in ggplot. The data I’m using is a small time series that we’ll use to plot a cumulative sum of via a line graph. It’s small enough to fit inline:<\/p>\n

library(ggplot2)\r\nlibrary(rbokeh)\r\nlibrary(htmlwidgets)\r\n\r\nstructure(list(wk = structure(c(16069, 16237, 16244, 16251, 16279,\r\n16286, 16300, 16307, 16314, 16321, 16328, 16335, 16342, 16349,\r\n16356, 16363, 16377, 16384, 16391, 16398, 16412, 16419, 16426,\r\n16440, 16447, 16454, 16468, 16475, 16496, 16503, 16510, 16517,\r\n16524, 16538, 16552, 16559, 16566, 16573), class = \"Date\"), n = c(1L,\r\n1L, 1L, 1L, 3L, 1L, 3L, 2L, 4L, 2L, 3L, 2L, 5L, 5L, 1L, 1L, 3L,\r\n3L, 3L, 2L, 1L, 1L, 1L, 2L, 1L, 2L, 7L, 1L, 2L, 6L, 7L, 1L, 1L,\r\n1L, 2L, 2L, 7L, 1L)), .Names = c(\"wk\", \"n\"), class = c(\"tbl_df\",\r\n\"tbl\", \"data.frame\"), row.names = c(NA, -38L)) -> by_week\r\n\r\nevents <- data.frame(when=as.Date(c(\"2014-10-09\", \"2015-03-20\", \"2015-05-15\")),\r\n                     what=c(\"Thing1\", \"Thing2\", \"Thing2\"))<\/pre>\n
The ggplot version is pretty straightforward:<\/p>\n
gg <- ggplot()\r\ngg <- gg + geom_vline(data=events,\r\n                      aes(xintercept=as.numeric(when), color=what),\r\n                      linetype=\"dashed\", alpha=1\/2)\r\ngg <- gg + geom_text(data=events,\r\n                     aes(x=when, y=1, label=what, color=what),\r\n                     hjust=1.1, size=3)\r\ngg <- gg + geom_line(data=by_week, aes(x=wk, y=cumsum(n)))\r\ngg <- gg + scale_x_date(expand=c(0, 0))\r\ngg <- gg + scale_y_continuous(limits=c(0, 100))\r\ngg <- gg + labs(x=NULL, y=\"Cumulative Stuff\")\r\ngg <- gg + theme_bw()\r\ngg <- gg + theme(panel.grid=element_blank())\r\ngg <- gg + theme(panel.border=element_blank())\r\ngg <- gg + theme(legend.position=\"none\")\r\ngg<\/pre>\n
We:<\/p>\n
- setup a base ggplot object
\n- add a layer of marker lines (which are the 3 `events` dates)
\n- add a layer of text for the marker lines
\n- add a layer of the actual line - note that we can use `cumsum(n)` vs pre-compute it
\n- setup scale and other aesthetic properties<\/p>\n
That gives us this:<\/p>\n
\"gg\"<\/a><\/p>\n
Here's a similar structure in rbokeh:<\/p>\n
figure(width=550, height=375,\r\n       logo=\"grey\", outline_line_alpha=0) %>%\r\n  ly_abline(v=events$when, color=c(\"red\", \"blue\", \"blue\"), type=2, alpha=1\/4) %>%\r\n  ly_text(x=events$when, y=5, color=c(\"red\", \"blue\", \"blue\"),\r\n          text=events$what, align=\"right\", font_size=\"7pt\") %>%\r\n  ly_lines(x=wk, y=cumsum(n), data=by_week) %>%\r\n  y_range(c(0, 100)) %>%\r\n  x_axis(grid=FALSE, label=NULL,\r\n         major_label_text_font_size=\"8pt\",\r\n         axis_line_alpha=0) %>%\r\n  y_axis(grid=FALSE,\r\n         label=\"Cumulative Stuff\",\r\n         minor_tick_line_alpha=0,\r\n         axis_label_text_font_size=\"10pt\",\r\n         axis_line_alpha=0) -> rb\r\nrb<\/pre>\n
Here, we set the `width` and `height` and configure some of the initial aesthetic options. Note that `outline_line_alpha=0` is the equivalent of `theme(panel.border=element_blank())`.<\/p>\n
The markers and text do not work exactly as one might expect since there's no way to specify a `data` parameter, so we have to set the colors manually. Also, since the target is a browser, points are specified in the same way you would with CSS. However, it's a pretty easy translation from `geom_[hv]line` to `ly_abline` and `geom_text` to `ly_text`.<\/p>\n
The `ly_lines` works pretty much like `geom_line`.<\/p>\n
Notice that both ggplot and rbokeh can grok dates for plotting (though we do not need the `as.numeric` hack for rbokeh).<\/p>\n
rbokeh will auto-compute bounds like ggplot would but I wanted the scale to go from 0 to 100 in each plot. You can think of `y_range` as `ylim` in ggplot.<\/p>\n
To configure the axes, you work directly with `x_axis` and `y_axis` parameters vs `theme` elements in ggplot. To turn off only lines, I set the alpha to 0 in each and did the same with the y axis minor tick marks. <\/p>\n
Here's the rbokeh result:<\/p>\n