I have a folium.Map that contains custom HTML Popups with clickable URLs. These Popups open when clicking on the polygons of the map. This is a feature that doesn't seem to be possible to achieve using holoviews.
My ideal example of the final application that I want to build with holoviews/geoviews is here with the source code here, but I would like to exchange the main map with my folium Map and plot polygons instead of rasterized points. Now when I would like to create the holoviews.DynamicMap from the folium.Map, holoviews complains (of course) that the data type "map" is not accepted. Is this somehow still possible?
I have found some notebook on GitHub where a holoviews plot in embedded in a folium map using a workaround that writes and reads again HTML, but it seems impossible to embed a folium map into holoviews such that other plots can be updated from this figure using Streams!?
Here is some toy data (from here) for the datasets that I use. For simplicity, let's assume I just had point data instead of polygons:
import folium as fn
def make_map():
m = fm.Map(location=[20.59,78.96], zoom_start=5)
green_p1 = fm.map.FeatureGroup()
green_p1.add_child(
fm.CircleMarker(
[row.Latitude, row.Longitude],
radius=10,
fill=True,
fill_color=fill_color,
fill_opacity=0.7
)
)
map.add_child(green_p1)
return map
If I understand it correctly, this needs to be tweaked now in the fashion that it can passed as the first argument to a holoviews.DynamicMap:
hv.DynamicMap(make_map, streams=my_streams)
where my_streams are some other plots that should be updated with the extent of the folium map.
Is that somehow possible or is my strategy wrong?
Related
Installed packages
holoviews 1.14.4, geoviews 1.9.1., matplotlib 3.4.2.
What I'm trying to do
I am trying to apply simple per-feature styles using GeoViews and the matplolib backend. I cannot figure out how to apply different edgecolor= parameters to different gv.Polygons elements in the same overlay. For some reason, they're always lightblue...
Similarly, facecolor= seems to have no effect.
Reproducible code sample
This uses a very small sample of the full dataset.
import pandas as pd
import geopandas as gpd
import geoviews as gv
from geoviews import opts
# loading both extensions as the full script calls for user input
# to choose between an interactive or static output
gv.extension('bokeh', 'matplotlib')
d1 = {'use': {0: 'Residential', 1: 'Residential'},
'geometry': {0: 'POLYGON ((13.80961103741604 51.04076975651729, 13.80965521888065 51.04079016168103, 13.80963851766593 51.04080454197601, 13.80959433642561 51.04078412781548, 13.80961103741604 51.04076975651729))',
1: 'POLYGON ((13.80977831740752 51.04313480566009, 13.80987122363639 51.04306085051974, 13.8099989591537 51.04312462457182, 13.80995486494384 51.04315973323087, 13.8099651184249 51.04316486464228, 13.80991634926543 51.04320371166482, 13.80977831740752 51.04313480566009))'}}
gdf1 = gpd.GeoDataFrame(pd.DataFrame(d1), geometry=gpd.GeoSeries.from_wkt(pd.DataFrame(d1)['geometry']), crs="EPSG:4326")
d2 = {'geometry': {1: 'POLYGON ((13.80894179055831 51.04544128170094, 13.80952887156242 51.0450399782091, 13.80954152432486 51.04504668985658, 13.80896834397535 51.04545611172818, 13.80894179055831 51.04544128170094))'}}
gdf2 = gpd.GeoDataFrame(pd.DataFrame(d2), geometry=gpd.GeoSeries.from_wkt(pd.DataFrame(d2)['geometry']), crs="EPSG:4326")
layout = gv.Polygons(gdf1, group="group1") * gv.Polygons(gdf2, group="group2")
layout.opts(
opts.Polygons('group1', cmap=['red'], edgecolor='black', linewidth=0.5, xaxis=None, yaxis=None, backend="matplotlib"),
opts.Polygons('group2', cmap=['lightblue'], edgecolor='blue', linewidth=0.5, backend="matplotlib"),
opts.Overlay(fig_size=500, backend='matplotlib')
)
gv.output(layout, backend='matplotlib')
gv.save(layout, "test.svg", dpi=600, backend='matplotlib')
Screenshot of the observed behaviour
This is a screen from the full dataset.
Expected behaviour
The red fill polygons belong to gdf1 and should have a black edgecolor but it's light blue instead. The blue fill polygon belongs to gdf2 and should have a lightblue fill and blue edgecolor, though the same color seems to be applied to both fill and edge.
What I've tried
Instead of using the group= parameter to specify styling for each of the Polygon elements (which I accidentally stumbled upon through the datashader documentation), I tried making multiple opts calls 'in-line' as suggested in the documentation for HoloViews here. This also has no effect.
Also, cmap=['color'] is the only method I've found to work to have GeoViews not use the automatically detected 'use' column in gdf1 as a vdim for color mapping. Is this the canonical approach and/or expected behaviour? color= or facecolor= seems to have no effect even though they are listed when calling gv.help(gv.opts.Polygons).
In short, I don't understand how to apply these particular styling parameters for the matplotlib backend and would very much appreciate any pointers.
2-Aug-21 Edit
Another strange behaviour seems to be that the figure in the, in my case VSCode-Python, interpreter, where the symbology seems to be faithfully represented, looks different from the .svg output generated by gv.save(layout, "test.svg", dpi=600, backend='matlplotlib'). The below images are outputs from the same run of the script.
Interpreter output:
gv.save() output:
How can I save a figure using PyPlot in Julia, so that the figure can be reloaded as a figure later in Julia? (not as an image)
You can use serialize to store any Julia object. This beautifully works for plots as well.
Let us start by generating a plot:
using Plots
pyplot()
p = plot(rand(10));
using Serialization
Serialization.serialize("myfile.jld", p);
Note that you need a semicolon after plot command so it does not appear on the screen.
Let us now read the plot (to have a full test I ended the previous Julia session and started a new one):
using Plots
pyplot();
using Serialization
p2 = Serialization.deserialize("myfile.jld");
In order to display it now it is enough to type in REPL:
julia> p2
You might want also want to use plain PyPlot (I strongly recommend Plots for flexibility). In that case your best bet is to follow rules described in object-oriented API of Matplotlib:
using PyPlot
ioff()
fig = subplot()
fig.plot(rand(10))
fig.set_title("Hello world")
using Serialization
serialize("pp.jld", fig)
In order to plot de-serialize back the object:
using PyPlot
ioff()
using Serialization
fig = deserialize("pp.jld")
show()
Finally, note that the serialization is good only for short term storage. If anything changes (e.g. you update Julia packages) you might not be able to de-serialize the plot.
Hence another good alternative for processable plots are saving them to LaTeX or SVG format - both is possible in Julia.
I'm mystified by how to use HoloViews styles to customize plots and achieve a consistent appearance across backends. HoloViews is billed as a package that provides an abstraction layer to several backends, notably Bokeh and Matplotlib, but I'm completely failing in my attempts to get plots generated using these backends to look the same. Settings in one backend are ignored by another, and each backend has many (most) formatting options missing, so that it is necessary to break through the abstraction to lower level calls directly the the backends.
I suspect I'm just missing something or have failed to discover the appropriate documentation.
The code below for example (using settings that don't attempt to produce the same appearance, but expose some of the issues) results in Matplotlib figures (right) that
ignore the attempt to achieve a uniform appearance for scatter plot point color,
ignore the attempt to override the color of histogram bars,
have marginal histograms with axis labels that are explicitly removed in the Bokeh versions (left),
have marginal histograms that are not framed and lack the vertical axis present in the Bokeh version,
have no control or customizations over the styling of axes, and
have additional subplot labels not present on the Bokeh plot.
In addition, there are many further customizations to both backend's plots (gridlines, frame color, for example) that I can't find settings for.
How do I set styles in HoloViews to achieve full and consistent control over plots produced by Bokeh and Matplotlib?
import numpy as np
import pandas as pd
import holoviews as hv
hv.extension('bokeh', 'matplotlib')
ds = hv.Dataset({'x': np.random.randn(100), 'y1': np.random.randn(100), 'y2': np.random.randn(100), 'y3': np.random.randn(100)},
['x'],['y1', 'y2', 'y3'])
def mpl_style_hook(plot, element):
# Settings required here are neither complete, nor do they correspond directly to the backend's naming
# Where is the correspondence between handles and the backend's names documented?
pass
def bok_style_hook(plot, element):
# Such a small set of abstractions is provided, it is almost always necessary to resort to hooks
plot.state.title.align = "center"
plot.handles['xaxis'].axis_label_text_color = 'red'
plot.handles['yaxis'].axis_label_text_color = 'green'
plot.handles['xaxis'].axis_label_text_font_style = "normal"
plot.handles['yaxis'].axis_label_text_font_style = "normal"
# Attempt to set options that apply to both backends; but ignored by Matplotlib
hv.opts.defaults(hv.opts.Scatter(color='green'), hv.opts.Histogram(fill_color='yellow'))
# Explictily set backend to avoid warnings (`backend=` isn't sufficient)
hv.Store.current_backend = 'bokeh'
hv.opts.defaults(
hv.opts.Scatter(line_color='orange', size=6, fill_alpha=1.0, hooks=[bok_style_hook]),
hv.opts.Histogram(fill_color='cyan', fill_alpha=0.9, line_width=1, line_color='gray', hooks=[bok_style_hook]),
backend='bokeh')
hv.Store.current_backend = 'matplotlib'
hv.opts.defaults(
hv.opts.Scatter(hooks=[mpl_style_hook]),
# Histogram color ignored
hv.opts.Histogram(color='orange', hooks=[mpl_style_hook]),
backend='matplotlib')
hv.Store.current_backend = 'bokeh'
s1 = hv.Scatter(ds, 'x', 'y1').opts(hv.opts.Scatter(labelled=[None, 'y'])).hist(num_bins=51, dimension=['x','y1'])
s2 = hv.Scatter(ds, 'x', 'y2').opts(hv.opts.Scatter(labelled=[None, 'y'])).hist(num_bins=51, dimension='y2')
s3 = hv.Scatter(ds, 'x', 'y3').hist(num_bins=51, dimension='y3')
p = (s1 + s2 + s3).opts(hv.opts.Histogram(labelled=[None, None]), hv.opts.Layout(shared_axes=True)).cols(1)
hv.save(p, '_testHV.html', backend='bokeh')
hv.save(p, '_testHV.png', backend='matplotlib')
p
I don't think you're missing anything in terms of actual software support; what you're missing is that HoloViews in no way promises to make it simple to make plots from different backends to look the same. The plots are meant to show the same data in roughly the same way, but the backends each work in different ways, and some of those differences are in fact reasons to choose that particular backend over another.
There are certainly ways that HoloViews could map from an abstract notion of styling into the details of how that's done in different backends, but that's surprisingly tricky. And very few users ask for that; most pick their favorite backend and just use it, and would rather we spend our limited development time working on other features.
That said, if the backends can produce similar plots, you should be able to work out settings for use with HoloViews that will generate them in matching form. To do this, you'd work out the settings one backend at a time, then apply them per backend. E.g. .opts(line_width=3, backend='bokeh').opts(linewidth=4.5, backend='matplotlib'), with the appropriate option being used when that object is displayed by each backend. Here the two options differ only by one character in their names, but they work very differently for such a seemingly simple concept of line width: matplotlib accepts a width in "points" (which depends on dpi and knowing the absolute size in inches), while bokeh accepts pixels in screen space. They are both widths, but there's not necessarily any direct way to compare the two values, as it depends on separate settings you may have done for dpi and fig_size. You should be able to get it to look similar with enough effort, but trying to achieve that across all plots for all time is a massive task that would need some separate funding and developers to achieve! Still, it's already much easier to do that in HoloViews than it would be to completely rewrite a plot between Matplotlib and Bokeh natively, so HoloViews is still helping a good bit, just not solving everything for you...
Plots is simple and powerful but sometimes I would like to have a little bit more control over individual elements of the plot to fine-tune its appearance.
Is it possible to update the plot object of the backend directly?
E.g., for the default pyplot backend, I tried
using Plots
p = plot(sin)
p.o[:axes][1][:xaxis][:set_ticks_position]("top")
but the plot does not change. Calling p.o[:show]() afterwards does not help, either.
In other words: Is there a way to use the PyPlot interface for a plot that was initially created with Plots?
Edit:
The changes to the PyPlot object become visible (also in the gui) when saving the figure:
using Plots
using PyPlot
p = Plots.plot(sin, top_margin=1cm)
gui() # not needed when using the REPL
gca()[:xaxis][:set_ticks_position]("top")
PyPlot.savefig("test.png")
Here, I used p.o[:axes][1] == gca(). One has to set top_margin=1cm because the plot area is not adjusted automatically (for my actual fine-tuning, this doesn't matter).
This also works for subsequent updates as long as only the PyPlot interface is used. E.g., after the following commands, the plot will have a red right border in addition to labels at the top:
gca()[:spines]["right"][:set_color]("red")
PyPlot.savefig("test.png")
However, when a Plots command like plot!(xlabel="foo") is used, all previous changes made with PyPlot are overwritten (which is not suprising).
The remaining question is how to update the gui interactively without having to call PyPlot.savefig explicitly.
No - the plot is a Plots object, not a PyPlot object. In your specific example you can do plot(sin, xmirror = true).
I'm trying to do the same but didn't find a solution to update an existing plot. But here is a partial answer: you can query information from the PyPlot axes object
julia> Plots.plot(sin, 1:4)
julia> Plots.PyPlot.plt[:xlim]()
(1.0,4.0)
julia> Plots.plot(sin, 20:24)
julia> ax = Plots.PyPlot.plt[:xlim]()
(20.0,24.0)
and it gets updated.
I have two different visualizations: a big plot (as PyPlot figure) and a worldmap (Compose context). Is there any possibility to combine those? (out of julia, without exporting both and then fiddling with some software).
I would like to do something like:
import PyPlot
using Compose
(fig, ax) = PyPlot.subplots(2,1)
ax[1][:plot]([1,2,3],[1,2,3])
composition = compose(compose(context(), circle()), fill("tomato"))
### magic command to add composition to ax[2]
In case there is no way to do it like this I will be greatefull for any other suggestions.
Thanks in advance!