gmtset BASEMAP_FRAME_RGB transparency - linux

I want to not have my basemap axes visible for an image I'm making in GMT with several plots in it. So far I have made them white using:
gmtset BASEMAP_FRAME_RGB = white
However some of the plots partially overlap and the white axes can be seen over the plots - any ideas on how to fix this? By making the basemap frame transparent ideally.

In fact I've realised I can change the order I do the plots in so that the overlap doesn't matter - also replaced -B... with -G255 to plot on a blank square.

Related

Holoviews points plot: NaN color setup

How do I set up the color for NaN in holoviews.points? Especially given that I am sampling the dataset with .where. Applying clipping colors gives the following error:
Unexpected option 'clipping_colors' for Overlay type across all extensions. No similar options found. It is not a background color either.
The code to generate my plot (bear in mind it is an xarray Dataset):
ds_30_365.beta.sel(pc=1).where(ds_30_365.pval.sel(pc=1)<0.2).hvplot.points(c='beta',coastline = True, cmap ='RdBu_r',clim=(-0.025,0.025)).opts(width=600,padding=(0,0))
The whole thing is an example of stippling and ideally I'd like to see ocean part being grey and the land remaining white. There is no data in the land part but everything ends up grey.
Thank you!

is it possible to display figures on a graph with matplotlib?

Is there a function in Matplotlib to display black rectangles on a graphic ?, like this :
You might want to look into Matplotlib's Patches and subsequent Rectangle modulus. Here's an example showcasing box creations with some error bars.

Using matplotlib.pyplot.clabel to add labels with a transparent background

For matplotlib.pyplot.clabel, is it possible to add labels to a contour plot without the white space behind the labels covering up any portion the contour lines? When I try to add labels to a contour plot it tends to cover up part of the contour plot with white space. Perhaps there is a way of making the background of the text transparent?
Use the inline argument
plt.clabel(..., inline=False)

How to draw a coordinate map containing both horizontal and vertical boxplot with matplotlib

I want to draw a picture like this using matplotlib.
How to do that?
Thank you in advance.
Example/i.stack.imgur.com/Vk57A.png
seaborn.jointplot might be what you're looking for. Specify to draw boxplots in the marginal plot and a regplot in the "main plot".
Seaborn builds on top of matplotlib, so you can still modify linecolors, markercolors etc. and draw annotations with plt.annotate.

Octave + Gnuplot rendering image upside down

I am displaying data using function imagesc(). If I set fltk as graphics_toolkit image is displayed correctly.
Can't post images directly(low reputation) http://i.stack.imgur.com/ARiwF.png
If I use gnuplot as plotting program image is rendered upside down.
fltk is unusable for me because its window isn't responding while function in octave is running. I also tried plot sine and it was correct plotted through fltk and gnuplot too.
There are at least two workarounds:
Use axis("xy") or axis("ij") to control the orientation of the y-axis, as decribed in the axis documentation
Use set(gca, "ydir", "normal") or set(gca, "ydir", "reverse"), as decribed in the axes properties doc

Resources