On Sat, 12 Aug 2023 17:40:00 -0700
American Citizen <[email protected]> wrote:
FYI.
My backend is set to TkAgg
I have it set explicitly in my .matplotlib/matplotlibrc file:
# the default backend; one of GTK GTKAgg GTKCairo CocoaAgg FltkAgg
# MacOSX QtAgg Qt4Agg TkAgg WX WXAgg Agg Cairo GDK PS PDF SVG Template
# You can also deploy your own backend outside of matplotlib by
# referring to the module name (which must be in the PYTHONPATH) as
# 'module://my_backend'
backend : TkAgg
don't believe i've ever tried to use another one, i've had tkagg set forever.
not even sure what the default is, but apparently it's not something useful.
What's very annoying is that if I try to change the backend it does not work,
it keeps usig TkAgg, and to make matters worse I've now found an additional
location for matplotlibrc and I'm not sure which one is correct.
config file hell is only slightly less annoying than dependency hell.
The web came through and
matplotlib.use("GTK3Cairo")
switched backends, and the plot time was no different than TKAgg.
I just ran a simple splot example which graphs '1/(x^2+y^2)' as a 2000 x 2000
surface and it shows up in < 1s.
A little laggy, on the order of 2 or 3s for 10000x10000.
At 20000x20000 it starts to get very slow. lots of cache misses probably . lol.
Here's the example.
https://matplotlib.org/stable/plot_types/3D/surface3d_simple.html
Brian
> I am using matplotlib, numpy, pandas all under python 3.6
>
> Now if I could just get the graphics backend under matplotlib to work.
> the current default rcParams['backend'] = 'agg' won't pop up any
> graphics window, I am forced to do a direct file write to save the image.
>
>
> On 8/12/23 06:03, Michael Ewan wrote:
> > Python and Pandas, then there are several graphing libraries to use, such
> > as Bokeh or Seaborn. Seaborn will give you great looking visualizations,
> > Bokeh adds an interactive interface.
> > https://seaborn.pydata.org/
> > http://bokeh.org/
> >
> > On Sat, Aug 12, 2023 at 5:42 AM Rich Shepard <[email protected]>
> > wrote:
> >
> >> On Fri, 11 Aug 2023, American Citizen wrote:
> >>
> >>> Has anyone used 3d surface plotting software (other than gnuplot splot
> >>> with pm3d) ??
> >> Randall,
> >>
> >> Three suggestions:
> >>
> >> 1. As Brian suggested, python3 and pandas. This would be my first choice
> >> for
> >> data sets such as yours for the shortest learning curve.
> >> <https://pandas.pydata.org/>
> >>
> >> 2. PSTricks. I've used this when precise control of vector graphics is
> >> required. <https://www.tug.org/PSTricks/main.cgi>
> >>
> >> 3. The R Project, the most comprehensive statistical data analysis toolkit
> >> available. Currently there are more than 6,000 packages in the
> >> Comprehensive
> >> R Archive Network covering every type of data, physical, chemical,
> >> biological, social, etc.
> >> <https://www.r-project.org; https://cran.r-project.org>
> >>
> >> Looks like a really interesting data analysis project. Good luch with it.
> >>
> >> Rich
> >>
> >>
--
Brian
