I am trying to use rgl knitr and webgl to plot several interactive 3d plots in a sequence.
And I am getting quite a weird behavior: when trying to rotate the plots with a mouse they seem to get back to their initial position after each mouse drag. This makes the interaction quite unintuitive. I noticed that the last plot on a page is not affected by this, but all the rest are.
This can be seen on the original knitr example:
source: https://dl.dropboxusercontent.com/u/15335397/misc/webgl-rmd.Rmd
plots: https://dl.dropboxusercontent.com/u/15335397/misc/webgl-rmd.html
Notice the differences when trying to rotate the first and second figure.
What could be a potential cause? Maybe the problem can be solved by passing some additional parameters to the 3d and rgl plots (was not able to find any)? Or is it a minor bug of knitr and webgl?
This is a bug in knitr, and has been fixed in the development version. The reason for it is that rgl includes <script src="CanvasMatrix.js" type="text/javascript"></script> for every plot, even when there are multiple plots on the same page. I have removed this line on knitr's side, so that CanvasMatrix.js is only loaded once. The knitr example you mentioned works now.
Related
Just started R a couple weeks ago. Getting familiar with ggplot. I successfully made the graph I wanted to including all legends, axes titles, etc. I'm able to export/save this image fine. However, the assignment also required me to change the font used for the title, axes, legend text, etc. on the graph. To import the font I wanted to use, I used showtext as I saw this recommended when I Googled. When I use this, it does change the font to what I want. However, although I can see the entire graphs and its two legends fine in the Plot viewer, when I go to export it, the legend text is now cut off on the right. This appears to be the case as soon as I load the showtext package and use showtext_auto(), even if I just try to then recreate the graph using R's default font, not even with the new font. I've spent tons of time playing around and can't seem to resolve this. Please help. Also would be preferable if there is a more permanent way to get fonts into R rather than me having to load showtext and whatever font I want each and every time. Thank you very much.
I want to implement interactive graph from plotly in overleaf. However, overleaf removed the communication with plotly in the V2 version. Is it still possible somehow to implement interactive graph in overleaf and therefore in pdf?
I want to hide/unhide line, zoom in and see value of different points when clicking on it.
I saw this question a couple times before, but the answere was only to zoom in or use javascript and implementing different picture on top of eachother.
However, this is not what i am looking for. I am looking for a really interactive in PDF. Does somebody know how this is possible?
I have a very complex plot, containing about 56,000 data points. It doesn't look right if I downsample it, so I really need to keep all of them. I would additionally like to add LaTeX captions to the figure. (The expression syntax, IMO, does not produce satisfactory rendering.)
After doing some digging around, TikZ seemed like the way to do it. But I found that it ran out of memory trying to plot the figure. I followed all of the advice I could find for TikZ memory management: this amounted to (1) using externalize and (2) increasing the main_memory for LaTeX to the maximum value (~12M). (I am using MacTeX 2014.) Neither of these solutions seemed to work.
At this point, having looked over SO and some other message boards, I am aware of only two options:
Switch to an alternate TeX interpreter, such as LuaTeX, which will allow me to use more memory, or
Use the native R plot, and then manually superimpose the desired labels onto the figures.
I consider (1) to be an acceptable solution, but the fact that I would need an alternate product makes me wonder if I am missing something. I wonder if there is a way to render complex native R plots, which happen to have TeX-style labels in them.
I'm trying to generate a 3d scatterplot using rgl. It looks great on my screen, but whenever I export it as a PDF (or any other postscript format) it completely ignores any size specifications I use.
(I'm running RGui v.2.15.1 and rgl v.0.92.892 on a Macbook under Mountain Lion.)
For example:
library(rgl)
set.seed(1982)
points3d(runif(5),runif(5),runif(5), size=20)
# points look huge
rgl.postscript('testplot.pdf', fmt='pdf')
# points look tiny
Does anyone have an idea for a way to get this to work? The resolution of the images I get using rgl.snapshot don't look so good, and I would really like to get a vector image for this plot.
Also, I followed this thread and I got text to resize just fine, but not points. So I thought one way to work around this would be to plot my points as text using a circle as my character, but I couldn't get rgl to accept symbols or expressions either...
Confirmed on Windows, look like some paper size scaling problem. You might try
spheres3d(runif(5),runif(5),runif(5),radius=0.1)
as a workaround if you can live with real 3d.
I know how to create a barplot, and how to stick it on a webpage; e.g, using hwriteImage in the hwriter package.
What I'd like is for each bar to be a region which highlights on mouseover, and where each bar has a different link when clicked. Similar to this map of the U.S. using the jQuery maphilight plugin, but for a barplot rather than a map. I imagine R could calculate the coordinates of the regions around each bar, generate the HTML AREA tag etc and pass this to maphilight quite easily. Has it been done already? I searched but no luck so far.
Have a look here, which summarises a couple of methods: rggobi and iplots. rggobi looks pretty promising, though maybe the installation looks a bit involved. iplots is only good for scatter plots.
Some other options (I think these are strongest ones at the moment):
googleVis
The googleVis package interfaces with the google charts API: try demo(googleVis) and the third & fourth one are bar chart (there could be more). It has the advantage of being pretty simple to get started with, although these are not R graphics:
df=data.frame(country=c("US", "GB", "BR"), val1=c(10,13,14), val2=c(23,12,32))
Column <- gvisColumnChart(df)
plot(Column)
gridSVG
The gridSVG exports the current grid graphics to an .svg file that can be included into a webpage. Unlike googleVis, it's R graphics (so you can use grid/ggplot2 which are more familiar). It looks like you may have to know some Javascript to further embellish your plots though (e.g. to animate on mouse over, you use grid.garnish(...,onmouseover=...)).
There's some example code you can try here (The really awesome ones are here - usually clicking on the "SVG file" link will have the full interactivity/animation.) (This one is a scatterplot where the points highlight when you move your mouse over them).
As I said - have a look at the package pages, demos, examples, etc to see which suits you.