Simple, concise geographical visualization in Python

Hello, I'm running Python 3.6.3 via Conda and I can not import GeoViews into a notebook. I've successfully run "conda install -c conda-forge -c ioam holoviews geoviews" in command line. But whenever I go to a notebook to import GeoViews I get the error below.

[1] Import GeoViews

ImportError Traceback (most recent call last) in () ----> 1 import geoviews

C:\Program Files\Anaconda3\lib\site-packages\geoviews_init_.py in () 7 import param 8 ----> 9 from .element import (_Element, Feature, Tiles, # noqa (API import) 10 WMTS, LineContours, FilledContours, Text, Image, 11 Points, Path, Polygons, Shape, Dataset, RGB,

C:\Program Files\Anaconda3\lib\site-packages\geoviews\element_init_.py in () 1 from holoviews.element import ElementConversion, Points as HvPoints 2 ----> 3 from .geo import (_Element, Feature, Tiles, is_geographic, # noqa (API import) 4 WMTS, Points, Image, Text, LineContours, RGB, 5 FilledContours, Path, Polygons, Shape, Dataset,

C:\Program Files\Anaconda3\lib\site-packages\geoviews\element\geo.py in () 3 from cartopy import crs as ccrs 4 from cartopy.feature import Feature as cFeature ----> 5 from cartopy.io.img_tiles import GoogleTiles as cGoogleTiles 6 from cartopy.io.shapereader import Reader 7 from holoviews.core import Element2D, Dimension, Dataset as HvDataset, NdOverlay

C:\Program Files\Anaconda3\lib\site-packages\cartopy\io\img_tiles.py in () 30 from future import (absolute_import, division, print_function) 31 ---> 32 from PIL import Image 33 import shapely.geometry as sgeom 34 import numpy as np

C:\Program Files\Anaconda3\lib\site-packages\PIL\Image.py in () 56 # Also note that Image.core is not a publicly documented interface, 57 # and should be considered private and subject to change. ---> 58 from . import _imaging as core 59 if PILLOW_VERSION != getattr(core, 'PILLOW_VERSION', None): 60 raise ImportError("The _imaging extension was built for another "

ImportError: DLL load failed: The specified module could not be found.

I attempted to make a Python 3.6 environment inclusive of datashader, holoviews, geoviews, and related dependencies. Here are the commands I ran:

conda create --name ds36 python=3.6
source activate ds36
conda install -c ioam/label/dev holoviews
conda install datashader
cd $C/datashader
python setup.py develop
conda install -c ioam paramnb
conda install -c scitools/label/dev -c conda-forge iris cartopy
conda install xarray
conda install -c conda-forge -c ioam holoviews>=1.6.2 geoviews

On the last line of that sequence of commands, I got the following error:

UnsatisfiableError: The following specifications were found to be in conflict:
  - geoviews -> python 2.7*
  - python 3.6*
Use "conda info <package>" to see the dependencies for each package.

Here's the ds36 at the time of that final failure:

appnope                   0.1.0                    py36_0
biggus                    0.15.0                   py36_0    conda-forge
bleach                    1.5.0                    py36_0
bokeh                     0.12.4                   py36_0
cartopy                   0.15.0              np112py36_0    conda-forge
cf_units                  1.1.3                    py36_1    conda-forge
chest                     0.2.3                    py36_0
cloudpickle               0.2.2                    py36_0
curl                      7.52.1                        0    conda-forge
cycler                    0.10.0                   py36_0
dask                      0.14.0                   py36_0
datashader                0.4.0                    py36_0
datashape                 0.5.4                    py36_0
decorator                 4.0.11                   py36_0
entrypoints               0.2.2                    py36_1
expat                     2.1.0                         2    conda-forge
freetype                  2.7                           1    conda-forge
geos                      3.5.1                         1    conda-forge
hdf4                      4.2.12                        0    conda-forge
hdf5                      1.8.17                        9    conda-forge
heapdict                  1.0.0                    py36_1
holoviews                 1.7dev6                  py36_0    ioam/label/dev
html5lib                  0.999                    py36_0
icu                       54.1                          0
ipykernel                 4.5.2                    py36_0
ipython                   5.3.0                    py36_0
ipython_genutils          0.1.0                    py36_0
ipywidgets                5.2.2                    py36_1
iris                      1.12.0                   py36_1    conda-forge
jbig                      2.1                           0
jinja2                    2.9.5                    py36_0
jpeg                      9b                            0
jsonschema                2.5.1                    py36_0
jupyter                   1.0.0                    py36_3
jupyter_client            5.0.0                    py36_0
jupyter_console           5.1.0                    py36_0
jupyter_core              4.3.0                    py36_0
libgfortran               3.0.0                         0    conda-forge
libiconv                  1.14                          4    conda-forge
libmo_unpack              3.0                           1    conda-forge
libnetcdf                 4.4.1.1                       2    conda-forge
libpng                    1.6.28                        0    conda-forge
libtiff                   4.0.6                         3
libxml2                   2.9.4                         0
libxslt                   1.1.29                        3    conda-forge
llvmlite                  0.16.0                   py36_0
locket                    0.2.0                    py36_1
lxml                      3.7.2                    py36_1    conda-forge
markupsafe                0.23                     py36_2
matplotlib                1.5.3               np112py36_7    conda-forge
mistune                   0.7.3                    py36_1
mkl                       2017.0.1                      0
mo_pack                   0.2.0               np112py36_0    conda-forge
multipledispatch          0.4.9                    py36_0
nbconvert                 5.1.1                    py36_0
nbformat                  4.3.0                    py36_0
nc_time_axis              1.0.1                    py36_0    conda-forge
netcdf4                   1.2.4               np112py36_3    conda-forge
networkx                  1.11                     py36_0
notebook                  4.3.1                    py36_0
numba                     0.31.0              np112py36_0
numpy                     1.12.0                   py36_0
odo                       0.5.0                    py36_1
olefile                   0.44                     py36_0
openssl                   1.0.2k                        0
owslib                    0.14.0                   py36_0    conda-forge
pandas                    0.19.2              np112py36_1
pandocfilters             1.4.1                    py36_0
param                     1.5.0                    py36_0    ioam
paramnb                   2.0.1                    py36_0    ioam
partd                     0.3.7                    py36_0
path.py                   10.1                     py36_0
pexpect                   4.2.1                    py36_0
pickleshare               0.7.4                    py36_0
pillow                    4.0.0                    py36_2    conda-forge
pip                       9.0.1                    py36_1
proj4                     4.9.3                         2    conda-forge
prompt_toolkit            1.0.9                    py36_0
ptyprocess                0.5.1                    py36_0
pyepsg                    0.3.1                    py36_0    conda-forge
pygments                  2.2.0                    py36_0
pyke                      1.1.1                    py36_0    conda-forge
pyparsing                 2.1.4                    py36_0
pyproj                    1.9.5.1                  py36_0    conda-forge
pyqt                      5.6.0                    py36_2
pyshp                     1.2.10                   py36_0    conda-forge
python                    3.6.0                         0
python-dateutil           2.6.0                    py36_0
pytz                      2016.10                  py36_0
pyyaml                    3.12                     py36_0
pyzmq                     16.0.2                   py36_0
qt                        5.6.2                         0
qtconsole                 4.2.1                    py36_1
readline                  6.2                           2
requests                  2.13.0                   py36_0
scipy                     0.18.1              np112py36_1
setuptools                27.2.0                   py36_0
shapely                   1.5.17              np112py36_2    conda-forge
simplegeneric             0.8.1                    py36_1
sip                       4.18                     py36_0
six                       1.10.0                   py36_0
sqlite                    3.13.0                        0
terminado                 0.6                      py36_0
testpath                  0.3                      py36_0
tk                        8.5.18                        0
toolz                     0.8.2                    py36_0
tornado                   4.4.2                    py36_0
traitlets                 4.3.2                    py36_0
udunits2                  2.2.23                        0    conda-forge
wcwidth                   0.1.7                    py36_0
wheel                     0.29.0                   py36_0
widgetsnbextension        1.2.6                    py36_0
xarray                    0.9.1                    py36_0
xz                        5.2.2                         1
yaml                      0.1.6                         0
zlib                      1.2.8                         3

I ran a similar sequence of commands for Python 2.7, 3.4, and 3.5 and the commands ran okay (I'm still now checking in those other Pythons envs if those envs actually work).

Datashader is very useful when working with geospatial data and holoviews now offers datashader operations. One major issue is that these operations do not currently handle a crs, which makes it difficult to work with them. The main obstacle here is that a) there is no geoviews RGB type, which could be used to hold the output of datashade and b) that projecting on every datashader plot update is very inefficient. I'd therefore suggest a temporary approach to make this work is to offer an operation that projects any Element type to a new crs and secondly to add a crs-aware RGB type to geoviews.

I have some bugs/issues when creating maps from shapefile. Here I use the UK referendum as an example to reproduce these issues.

(1) If there are missing values, the corresponding spatial unit on the map will not be displayed. it would be better to have an option to automatically fill the missing value to display all the spatial unit even though they don't .

(2)If the data type is integer. The map will not show color correctly.

(3) When the data value contains nan, the map looks good. But when these nan values are filled by 0. the resulting map is totally incorrect. (Here I manually assign three nan and fill them by zero)

(4) Latitude and longitude on the map also seem to be problematic.

(5) Better to have way to set color range for these maps. I am affected mostly by the third issue, because in my study I have to fill some values, otherwise it will become the first issue.

Hope these issues can be fixed or improved.

Code to reproduce these issues is attached. I don't know how to upload jupyter notebook.

import pandas as pd
import numpy as np
import holoviews as hv
import geoviews as gv
import cartopy
from cartopy import crs as ccrs
%pylab

hv.notebook_extension('bokeh','matplotlib')

shapefile = 'boundaries/boundaries.shp'
shapes = cartopy.io.shapereader.Reader(shapefile)

referendum = pd.read_csv('referendum.csv')
referendum['value1'] = referendum['leaveVoteshare']

%%output backend='bokeh'
%%opts Shape (cmap='viridis') [xaxis=None yaxis=None tools=['hover'] width=400 height=500]

# Case 1: set referendum value is interger
referendum_modified = referendum.copy()
referendum_modified['value1'] = referendum_modified['value1'].apply(np.int)

data = hv.Dataset(referendum_modified)

gv.Shape.from_records(shapes.records(), data, on='code', value='value1',
                     index=['name', 'regionName'], crs=ccrs.PlateCarree())

%%output backend='bokeh'
%%opts Shape (cmap='viridis') [xaxis=None yaxis=None tools=['hover'] width=400 height=500]

# Case 2: If referendem values are missing for some counties
referendum_modified = referendum.copy()

referendum_modified = referendum_modified.iloc[0:100,:]

data = hv.Dataset(referendum_modified)

gv.Shape.from_records(shapes.records(), data, on='code', value='value1',
                     index=['name', 'regionName'], crs=ccrs.PlateCarree())

%%output backend='bokeh'
%%opts Shape (cmap='viridis') [xaxis=None yaxis=None tools=['hover'] width=400 height=500]

# Case 3: Here set the fist three rows to nan and fill them with 0   
referendum_modified = referendum.copy()
referendum_modified['value1'][0:3]=np.nan
referendum_modified['value1'] = referendum_modified['value1'].fillna(0)

data = hv.Dataset(referendum_modified)

    
gv.Shape.from_records(shapes.records(), data, on='code', value='value1',
                     index=['name', 'regionName'], crs=ccrs.PlateCarree())

I'm visualizing public Landsat data with hvplot.xarray but I think this is likely a geoviews issue.

It takes about 3 seconds to render an image when not using the crs:

It takes about 3 minutes to render an image when using the crs:

Without crs it's using my dask cluster, but with crs it does not use the cluster.

This is the code I'm using:

import rasterio 
import xarray as xr
import hvplot.xarray
from dask.distributed import Client, progress
import cartopy.crs as ccrs

client = Client()

image_url = 'https://storage.googleapis.com/gcp-public-data-landsat/LC08/01/047/027/LC08_L1TP_047027_20130421_20170310_01_T1/LC08_L1TP_047027_20130421_20170310_01_T1_B4.TIF'

da = xr.open_rasterio(image_url, chunks={'band': 1, 'x': 2048, 'y': 2048})

band1 = da.sel(band=1).persist()

display(band1.hvplot(rasterize=True, width=600, height=400, cmap='viridis'))

crs = ccrs.UTM(zone='10n')
display(band1.hvplot(crs=crs, rasterize=True, width=600, height=400, cmap='viridis'))

Here is the notebook.

Is this expected?

Is there anything I can do to speed up visualization with crs specified?

Adds a GeoPandas interface allowing Path and Polygon types to wrap around geopandas dataframes directly. Here is a basic example:

https://anaconda.org/philippjfr/geopandas/notebook

Requires holoviews master and will likely be released at the same time as HoloViews 1.9.

I'm trying to install geoviews on a linux machine. I created a new conda environment to try it out and ran the following commands successfully:

conda create --name geoviews jupyter
conda install -c scitools/label/dev -c conda-forge iris cartopy
conda install xarray
conda install -c ioam holoviews

However when trying to install geoviews I found a problem:

conda install -c ioam geoviews
Fetching package metadata: ......
Solving package specifications: ....................................................
Error: Could not find some dependencies for geoviews: cartopy

I can confirm that cartopy is installed:

conda list
...
cartopy                   0.14.2              np111py27_1    conda-forge
...

so I'm not sure what the problem is?

I have a GDF that I've created by overlaying a few other GDF's (state district data for various offices). I want to plot that on top of tile data.

Plotting the GDF with matplotlib via geopandas seems to work OK:

but creating a Polygons object seems to go wrong:

My ultimate goal is a colorized map with a hover tool: each district's color represents a unique combination of districts that cover that area, and the hover tool will tell you which districts cover a given area. In Travis County, there are 58 such combinations, so a legend is not really feasible. Color-coding would still be nice, though, to make the separations more clear.

I'm happy to supply the GDF - please suggest the best way to do so.

https://ioam.github.io/geoviews/ is built on pushing to this PR.

To do:

• [x] Fill in placeholder content, making coherent
• [x] ensure haven't removed anything from existing site

Before merging:

• [x] squash
• [x] update travis file to e.g. build site on tag only

Since the recent upgrade to Holoviews=1.13.1, Geoviews=1.8..1 and Bokeh=2.0.0, I have frequent trouble to create and pinning envs with these packages.

The following somehow works:

conda install -c pyviz -c conda-forge geopandas fiona jupyterlab geoviews holoviews nbconvert descartes numpy mapclassify bokeh jupyter_contrib_nbextensions

but it will install Geoviews 1.6.6 and pyproj=2.6.0.

This also doesn't work:

conda install -c pyviz -c conda-forge geopandas fiona jupyterlab "geoviews=1.8.1" holoviews nbconvert descartes numpy mapclassify bokeh jupyter_contrib_nbextensions

It will install pyproj=1.9.6, it also falls back to flexible solve and takes really long.

I was wondering if there's an explanation for this behavior? How can I pin geoviews to 1.8.1 from the pyviz channel?

When using -c pyviz/label/dev, conda also falls back to flexible solve. The results seem to change pretty much every day.

I've tried with fewer packages, but still - solving the environment takes very long and it almost always falls back to flexible solve (my env is set to strict).

Here's the complete order of steps to reproduce:

conda create -n geoviews_env python=3.7 -c pyviz -c conda-forge
conda activate geoviews_env 
conda config --env --set channel_priority strict
conda config --show channel_priority # verify
conda install -c pyviz -c conda-forge geopandas fiona jupyterlab geoviews holoviews nbconvert descartes numpy mapclassify bokeh jupyter_contrib_nbextensions

The documentation is very short when it comes to installation instructions, but I often encounter problems like these. Ideally, I want to provide a guide to the readers of my notebooks so that they can re-create an env that will be guaranteed compatible to the code in the Notebook.

• running conda 4.8.3 in Linux (WSL)

I have a local jupyter lab instance where this code works just fine:

import geoviews as gv
import cartopy.crs as ccrs
gv.extension('bokeh')

poly = [
    (-95.777304976205, 29.70081935868131),
    (-95.777131944776, 29.70009523405696),
    (-95.780067725666, 29.69971579007273),
    (-95.780184253531, 29.70061729352786),
    (-95.778716019600, 29.70071746836681),
    (-95.777304976205, 29.70081935868131)
]
gv.Polygons([poly], crs=ccrs.PlateCarree())

Then I have a JupyterHub instance with the same holoviz packages, but the above code produces a warning about the projection being wrong and results in an empty plot. While projecting a Polygons element from a PlateCarree coordinate reference system (crs) to a Mercator projection none of the projected paths were contained within the bounds specified by the projection.

If I add the projection to .opts, e.g.:
gv.Polygons([poly], crs=ccrs.PlateCarree()).opts(projection=ccrs.PlateCarree())
it will work properly on JHub. Any idea why the extra projection specification is required for my JHub instance?

geoviews=1.6.4 from conda-forge

With the release of numpy 1.24, the warning for inhomogeneous shape is now a ValueError.

For Geoviews to work with the new numpy version, https://github.com/holoviz/holoviews/pull/5581 is also needed.

TODO

• [x] Need to check if examples can run

This PR does two thing

1. Switch to holoviz_tasks/install.
2. Remove doit_test_flakes as the pre-commit already does this

I have pinned numpy because version 1.24 now raises an error when an inhomogeneous shape is passed to np.array/np.asarray

Software version info:

geoviews=1.9.5
holoviews=1.15.3
bokeh=2.4.3
python=3.8.15
jupyterlab=3.5.2
shapely=2.0.0

I can plot every other feature geometry just fine in my notebook but just gf.states as a singular command in its own cell throws an error from trying to iterate through a MultiPolygon:

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/IPython/core/formatters.py:972, in MimeBundleFormatter.__call__(self, obj, include, exclude)
    969     method = get_real_method(obj, self.print_method)
    971     if method is not None:
--> 972         return method(include=include, exclude=exclude)
    973     return None
    974 else:

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/holoviews/core/dimension.py:1293, in Dimensioned._repr_mimebundle_(self, include, exclude)
   1286 def _repr_mimebundle_(self, include=None, exclude=None):
   1287     """
   1288     Resolves the class hierarchy for the class rendering the
   1289     object using any display hooks registered on Store.display
   1290     hooks.  The output of all registered display_hooks is then
   1291     combined and returned.
   1292     """
-> 1293     return Store.render(self)

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/holoviews/core/options.py:1426, in Store.render(cls, obj)
   1424 data, metadata = {}, {}
   1425 for hook in hooks:
-> 1426     ret = hook(obj)
   1427     if ret is None:
   1428         continue

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/holoviews/ipython/display_hooks.py:277, in pprint_display(obj)
    275 if not ip.display_formatter.formatters['text/plain'].pprint:
    276     return None
--> 277 return display(obj, raw_output=True)

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/holoviews/ipython/display_hooks.py:247, in display(obj, raw_output, **kwargs)
    245 elif isinstance(obj, (CompositeOverlay, ViewableElement)):
    246     with option_state(obj):
--> 247         output = element_display(obj)
    248 elif isinstance(obj, (Layout, NdLayout, AdjointLayout)):
    249     with option_state(obj):

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/holoviews/ipython/display_hooks.py:141, in display_hook.<locals>.wrapped(element)
    139 try:
    140     max_frames = OutputSettings.options['max_frames']
--> 141     mimebundle = fn(element, max_frames=max_frames)
    142     if mimebundle is None:
    143         return {}, {}

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/holoviews/ipython/display_hooks.py:187, in element_display(element, max_frames)
    184 if type(element) not in Store.registry[backend]:
    185     return None
--> 187 return render(element)

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/holoviews/ipython/display_hooks.py:68, in render(obj, **kwargs)
     65 if renderer.fig == 'pdf':
     66     renderer = renderer.instance(fig='png')
---> 68 return renderer.components(obj, **kwargs)

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/holoviews/plotting/renderer.py:398, in Renderer.components(self, obj, fmt, comm, **kwargs)
    395 embed = (not (dynamic or streams or self.widget_mode == 'live') or config.embed)
    397 if embed or config.comms == 'default':
--> 398     return self._render_panel(plot, embed, comm)
    399 return self._render_ipywidget(plot)

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/holoviews/plotting/renderer.py:405, in Renderer._render_panel(self, plot, embed, comm)
    403 doc = Document()
    404 with config.set(embed=embed):
--> 405     model = plot.layout._render_model(doc, comm)
    406 if embed:
    407     return render_model(model, comm)

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/panel/viewable.py:508, in Renderable._render_model(self, doc, comm)
    506 if comm is None:
    507     comm = state._comm_manager.get_server_comm()
--> 508 model = self.get_root(doc, comm)
    510 if config.embed:
    511     embed_state(self, model, doc,
    512                 json=config.embed_json,
    513                 json_prefix=config.embed_json_prefix,
    514                 save_path=config.embed_save_path,
    515                 load_path=config.embed_load_path,
    516                 progress=False)

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/panel/viewable.py:559, in Renderable.get_root(self, doc, comm, preprocess)
    542 """
    543 Returns the root model and applies pre-processing hooks
    544 
   (...)
    556 Returns the bokeh model corresponding to this panel object
    557 """
    558 doc = init_doc(doc)
--> 559 root = self._get_model(doc, comm=comm)
    560 if preprocess:
    561     self._preprocess(root)

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/panel/layout/base.py:146, in Panel._get_model(self, doc, root, parent, comm)
    144 if root is None:
    145     root = model
--> 146 objects = self._get_objects(model, [], doc, root, comm)
    147 props = dict(self._init_params(), objects=objects)
    148 model.update(**self._process_param_change(props))

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/panel/layout/base.py:131, in Panel._get_objects(self, model, old_objects, doc, root, comm)
    129 else:
    130     try:
--> 131         child = pane._get_model(doc, root, model, comm)
    132     except RerenderError:
    133         return self._get_objects(model, current_objects[:i], doc, root, comm)

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/panel/pane/holoviews.py:265, in HoloViews._get_model(self, doc, root, parent, comm)
    263     plot = self.object
    264 else:
--> 265     plot = self._render(doc, comm, root)
    267 plot.pane = self
    268 backend = plot.renderer.backend

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/panel/pane/holoviews.py:342, in HoloViews._render(self, doc, comm, root)
    339     if comm:
    340         kwargs['comm'] = comm
--> 342 return renderer.get_plot(self.object, **kwargs)

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/holoviews/plotting/bokeh/renderer.py:70, in BokehRenderer.get_plot(self_or_cls, obj, doc, renderer, **kwargs)
     63 @bothmethod
     64 def get_plot(self_or_cls, obj, doc=None, renderer=None, **kwargs):
     65     """
     66     Given a HoloViews Viewable return a corresponding plot instance.
     67     Allows supplying a document attach the plot to, useful when
     68     combining the bokeh model with another plot.
     69     """
---> 70     plot = super().get_plot(obj, doc, renderer, **kwargs)
     71     if plot.document is None:
     72         plot.document = Document() if self_or_cls.notebook_context else curdoc()

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/holoviews/plotting/renderer.py:240, in Renderer.get_plot(self_or_cls, obj, doc, renderer, comm, **kwargs)
    237     defaults = [kd.default for kd in plot.dimensions]
    238     init_key = tuple(v if d is None else d for v, d in
    239                      zip(plot.keys[0], defaults))
--> 240     plot.update(init_key)
    241 else:
    242     plot = obj

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/holoviews/plotting/plot.py:948, in DimensionedPlot.update(self, key)
    946 def update(self, key):
    947     if len(self) == 1 and ((key == 0) or (key == self.keys[0])) and not self.drawn:
--> 948         return self.initialize_plot()
    949     item = self.__getitem__(key)
    950     self.traverse(lambda x: setattr(x, '_updated', True))

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/geoviews/plotting/bokeh/plot.py:111, in GeoPlot.initialize_plot(self, ranges, plot, plots, source)
    109 def initialize_plot(self, ranges=None, plot=None, plots=None, source=None):
    110     opts = {} if isinstance(self, HvOverlayPlot) else {'source': source}
--> 111     fig = super(GeoPlot, self).initialize_plot(ranges, plot, plots, **opts)
    112     if self.geographic and self.show_bounds and not self.overlaid:
    113         from . import GeoShapePlot

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/holoviews/plotting/bokeh/element.py:1397, in ElementPlot.initialize_plot(self, ranges, plot, plots, source)
   1394     self.handles['y_range'] = plot.y_range
   1395 self.handles['plot'] = plot
-> 1397 self._init_glyphs(plot, element, ranges, source)
   1398 if not self.overlaid:
   1399     self._update_plot(key, plot, style_element)

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/holoviews/plotting/bokeh/element.py:1341, in ElementPlot._init_glyphs(self, plot, element, ranges, source)
   1339 else:
   1340     style = self.style[self.cyclic_index]
-> 1341     data, mapping, style = self.get_data(element, ranges, style)
   1342     current_id = element._plot_id
   1344 with abbreviated_exception():

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/geoviews/plotting/bokeh/__init__.py:237, in FeaturePlot.get_data(self, element, ranges, style)
    235     el_type = Polygons
    236 polys = el_type(geoms, crs=element.crs, **util.get_param_values(element))
--> 237 return super(FeaturePlot, self).get_data(polys, ranges, style)

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/geoviews/plotting/bokeh/plot.py:170, in GeoPlot.get_data(self, element, ranges, style)
    168 def get_data(self, element, ranges, style):
    169     if self._project_operation and self.geographic:
--> 170         element = self._project_operation(element, projection=self.projection)
    171     return super(GeoPlot, self).get_data(element, ranges, style)

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/param/parameterized.py:3658, in ParameterizedFunction.__new__(class_, *args, **params)
   3656 inst = class_.instance()
   3657 inst.param._set_name(class_.__name__)
-> 3658 return inst.__call__(*args,**params)

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/holoviews/core/operation.py:220, in Operation.__call__(self, element, **kwargs)
    218 kwargs['link_dataset'] = self._propagate_dataset
    219 kwargs['link_inputs'] = self.p.link_inputs
--> 220 return element.apply(self, **kwargs)

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/holoviews/core/accessors.py:45, in AccessorPipelineMeta.pipelined.<locals>.pipelined_call(*args, **kwargs)
     42     inst._obj._in_method = True
     44 try:
---> 45     result = __call__(*args, **kwargs)
     47     if not in_method:
     48         init_op = factory.instance(
     49             output_type=type(inst),
     50             kwargs={'mode': getattr(inst, 'mode', None)},
     51         )

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/holoviews/core/accessors.py:202, in Apply.__call__(self, apply_function, streams, link_inputs, link_dataset, dynamic, per_element, **kwargs)
    200 if hasattr(apply_function, 'dynamic'):
    201     inner_kwargs['dynamic'] = False
--> 202 new_obj = apply_function(self._obj, **inner_kwargs)
    203 if (link_dataset and isinstance(self._obj, Dataset) and
    204     isinstance(new_obj, Dataset) and new_obj._dataset is None):
    205     new_obj._dataset = self._obj.dataset

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/holoviews/core/operation.py:214, in Operation.__call__(self, element, **kwargs)
    210         return element.clone([(k, self._apply(el, key=k))
    211                               for k, el in element.items()])
    212     elif ((self._per_element and isinstance(element, Element)) or
    213           (not self._per_element and isinstance(element, ViewableElement))):
--> 214         return self._apply(element)
    215 elif 'streams' not in kwargs:
    216     kwargs['streams'] = self.p.streams

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/holoviews/core/operation.py:141, in Operation._apply(self, element, key)
    139     if not in_method:
    140         element._in_method = True
--> 141 ret = self._process(element, key)
    142 if hasattr(element, '_in_method') and not in_method:
    143     element._in_method = in_method

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/geoviews/operation/projection.py:40, in _project_operation._process(self, element, key)
     39 def _process(self, element, key=None):
---> 40     return element.map(self._process_element, self.supported_types)

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/holoviews/core/data/__init__.py:204, in PipelineMeta.pipelined.<locals>.pipelined_fn(*args, **kwargs)
    201     inst._in_method = True
    203 try:
--> 204     result = method_fn(*args, **kwargs)
    205     if PipelineMeta.disable:
    206         return result

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/holoviews/core/data/__init__.py:1216, in Dataset.map(self, *args, **kwargs)
   1214 @wraps(LabelledData.map)
   1215 def map(self, *args, **kwargs):
-> 1216     return super(Dataset, self).map(*args, **kwargs)

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/holoviews/core/dimension.py:700, in LabelledData.map(self, map_fn, specs, clone)
    698     return deep_mapped
    699 else:
--> 700     return map_fn(self) if applies else self

File /storage/work/c/cnd5285/mambaforge/envs/uviz/lib/python3.8/site-packages/geoviews/operation/projection.py:79, in project_path._process_element(self, element)
     77     continue
     78 elif isinstance(geom, MultiPolygon):
---> 79     polys = [g for g in geom if g.area > 1e-15]
     80     if not polys:
     81         continue

TypeError: 'MultiPolygon' object is not iterable

Fixes #592

References:

• https://shapely.readthedocs.io/en/latest/release/2.x.html
• https://shapely.readthedocs.io/en/latest/migration_pygeos.html
• https://github.com/SciTools/iris/issues/3528
• https://github.com/pangeo-data/xESMF/pull/214

ALL software version info

GeoViews 1.9.5, HoloViews 1.15.1 on Linux

The code provided below is executed in a notebook (jupyter_client 7.0.6, jupyter_core 4.8.1, jupyter_server 1.11.1, notebook 6.4.4) but I'm getting the same bug in a server (Panel 0.14.0).

This is not a regression as I've experienced this issue for more than a year with various versions.

Description of expected behavior and the observed behavior

When working in geostationary projection (or orthographic and presumably others) with Image or RGB elements, it is preferable to provide bounds calculated in the projection coordinates instead of lat/lon. This is because the corners of the element may not fit within the globe and therefore may not correspond to a valid (lat, lon) pair:

When the image is large, regridding using hd.regrid allows to drastically reduce rendering time. Likewise, using resample_geometry on Feature objects (coastline typically) also helps.

However you can't use both at the same time. If you do, the plot does render, but the moment you pan or zoom, one of the element disappears (geometry or image, depending on their order in the overlay). The same thing happens if you use hv.Image/hv.RGB instead of their gv counterparts, which I do (it's a bit faster).

So in this case you have to choose between image regridding and feature resampling. Apparently the bug does not occur if you provide bounds in lat/lon.

Complete, minimal, self-contained example code that reproduces the issue

import numpy as np
import geoviews as gv
import holoviews.operation.datashader as hd
import cartopy.crs as ccrs

gv.extension('bokeh')

coastline = gv.feature.coastline.geoms('10m').opts(color='black')
resampled_coastline = gv.operation.resample_geometry(coastline)

projection = ccrs.Orthographic()
opts = dict(projection=projection, width=400, height=400)

ls = np.linspace(0, 10, 2000)
xx, yy = np.meshgrid(ls, ls)

# In separate cells:
latlon_bounds = (-20, -10, 20, 10)
latlon_img = gv.Image(np.sin(2*xx)*np.cos(yy), bounds=latlon_bounds)
regrid_latlon_img = hd.regrid(latlon_img)
(regrid_latlon_img*resampled_coastline).opts(**opts) # Works fine

# and
geo_bounds = (-2000000, -1000000, 2000000, 1000000)
geo_img = gv.Image(np.sin(2*xx)*np.cos(yy), bounds=geo_bounds, crs=projection)
# or geo_img = hv.Image(np.sin(2*xx)*np.cos(yy), bounds=geo_bounds)
regrid_geo_img = hd.regrid(geo_img)
(regrid_geo_img*resampled_coastline).opts(**opts) # Coastline will disappear as soon as you pan or zoom

Stack traceback and/or browser JavaScript console output

Nothing is logged in standard output or in the browser console

Screenshots or screencasts of the bug in action

https://user-images.githubusercontent.com/49269671/195110736-d1e23249-6e91-4859-8268-ecb1dead9bb1.mp4