Currently the polling_interval is only used as an initial configuration and can not be dynamically changed.

The following is a discourse link where I have asked the question whether this is a supported feature.
https://discourse.bokeh.org/t/dynamically-changing-bokeh-ajaxdatasource-polling-interval/7737

When plotting plt.plot(np.ones(10), np.ones((10,0)) it raises a ZeroDivisionError, which confused me much.

Code for reproduction

import matplotlib.pyplot as plt
import numpy as np

plt.plot(np.ones(10), np.ones((10,0)))

This raises the error:

ZeroDivisionError: integer division or modulo by zero

Expected outcome

I think however, it should either r

it's becoming more time-consuming and error-prone to manually re-test all the demos following internal refactorings and API adjustments.

now that the API is fleshed out a bit, it's possible to test a large amount of code (non-granularly) without having to simulate all interactions via Puppeteer or similar.

a lot of code can already be regression-tested by simply running all the demos and val

https://readthedocs.org/projects/chartify/builds/

What's wrong?

It is currently a pain to use an OxyPlot.WinfowsForms.PlotView with a transparent background: it throws if you set its BackColor to transparent. To avoid this, it is necessary to set the ControlStyles.SupportsTransparentBackColor style to true on the PlotView; however, Control.SetStyle is protected, so consumers must resort to reflection or extending PlotView to do so. This could be

I think it could be useful, when one wants to plot only e.g. class 1, to have an option to produce consistent plots for both plot_cumulative_gain and plot_roc

At the moment, instead, only plot_roc supports such option.

Thanks a lot

Michael Bostock has made some major improvements to the overlap avoidance algorithms in d3-sankey over the last week. Since the code in holoviews is a fairly straightforward Python port of that code it should be reasonably straightforward to update our code to match. Here's a link to the diff for all the updates.

I don't think this is a duplicate ticket, even though it's quite an obvious feature after that cat demo at SciPy2018.

The code that selects and renders the different geos is here in the THREE.js code
https://github.com/maartenbreddels/ipyvolume/blob/master/js/src/scatter.js#L57

I don't know quite where/how to inject the mesh definition compatible with the data file of this form, but i

What is the feature ?

I'm trying out this library for the first time and I'm seeing several places where some shorthand functions would have been really pleasant:

1. Shorthand functions for Pos: Pos::top_right, Pos::bottom_right, Pos::bottom_center etc..
2. Short way to set colors for components. My plots are supposed to be white-on-black, which requires a lot of boilerplate on lots o

Details

(This is related to JuliaPlots/Plots.jl#1706)

For example scatter(skipmissing([1.0, 2.0, missing, 1.0])) fails with Cannot convert Base.SkipMissing{Vector{Union{Missing, Float64}}} to series data for plotting.

Generally skipmissing is not required but it seems odd to see this fail.

Also I stumbled on this while trying to make a histogram, e.g.

I've noticed that calling delete! on an Axis3 does not completely remove it:

julia> fig = Figure()

julia> ax = Axis3(fig[1, 1])
Axis3()

julia> surface!(ax, rand(100, 100))
Surface{Tuple{IntervalSets.ClosedInterval{Float32}, IntervalSets.ClosedInterval{Float32}, Matrix{Float32}}}

julia> delete!(ax)

![frame](https://user-images.githubusercontent.com/6333339/115960

there is no feature to add control MarkerShape in PlotRadar!
it would be cool and interesting to add this feature.

So as we start to see 1Hz soundings in the wild, we start to see issues with our calculations and the fact that sometimes pressure, as directly reported, is not necessarily monotonically decreasing. For instance, see this
sample sounding. There are a couple of spots where the pressure bounces up for a single point by 0.1 h