Setting levels for 2D KDE contour plot returns errors for Bokeh backend

Describe the bug Setting the levels parameter for either contour_kwargs or contourf_kwargs in plot_kde returns errors for the Bokeh backend.

To Reproduce

import numpy as np
import arviz as az

rng = np.random.default_rng()
data = rng.multivariate_normal([2, 2], [[1, 0.4], [0.4, 0.8]], 1000000)

levels = [0, 0.01, 0.07, 0.13, 0.2]
ax = az.plot_kde(
    data[:, 0],
    data[:, 1],
    contourf_kwargs={"levels": levels},
    contour_kwargs={"levels": levels},
    backend="bokeh",
)

returns the error

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-7-6f97bf89ac37> in <module>
      6 
      7 levels = [0, 0.01, 0.07, 0.13, 0.2]
----> 8 ax = az.plot_kde(
      9     data[:, 0],
     10     data[:, 1],

/opt/arviz/arviz/plots/kdeplot.py in plot_kde(values, values2, cumulative, rug, label, bw, adaptive, circular, quantiles, rotated, contour, hdi_probs, fill_last, figsize, textsize, plot_kwargs, fill_kwargs, rug_kwargs, contour_kwargs, contourf_kwargs, pcolormesh_kwargs, is_circular, ax, legend, backend, backend_kwargs, show, return_glyph, **kwargs)
    341     # TODO: Add backend kwargs
    342     plot = get_plotting_function("plot_kde", "kdeplot", backend)
--> 343     ax = plot(**kde_plot_args)
    344 
    345     return ax

/opt/arviz/arviz/plots/backends/bokeh/kdeplot.py in plot_kde(density, lower, upper, density_q, xmin, xmax, ymin, ymax, gridsize, values, values2, rug, label, quantiles, rotated, contour, fill_last, figsize, textsize, plot_kwargs, fill_kwargs, rug_kwargs, contour_kwargs, contourf_kwargs, pcolormesh_kwargs, is_circular, ax, legend, backend_kwargs, show, return_glyph)
    185             else:
    186                 levels_scaled_nonclip = _scale_axis(np.asarray(levels), scaled_density_args)
--> 187                 levels_scaled = np.clip(levels_scaled_nonclip, 0, 1)
    188 
    189             cmap = contourf_kwargs.pop("cmap", "viridis")

<__array_function__ internals> in clip(*args, **kwargs)

/usr/local/envs/testenv_3.8_PYSTAN_latest_PYRO_latest_EMCEE_latest_TF_latest/lib/python3.8/site-packages/numpy/core/fromnumeric.py in clip(a, a_min, a_max, out, **kwargs)
   2095 
   2096     """
-> 2097     return _wrapfunc(a, 'clip', a_min, a_max, out=out, **kwargs)
   2098 
   2099 

/usr/local/envs/testenv_3.8_PYSTAN_latest_PYRO_latest_EMCEE_latest_TF_latest/lib/python3.8/site-packages/numpy/core/fromnumeric.py in _wrapfunc(obj, method, *args, **kwds)
     53     bound = getattr(obj, method, None)
     54     if bound is None:
---> 55         return _wrapit(obj, method, *args, **kwds)
     56 
     57     try:

/usr/local/envs/testenv_3.8_PYSTAN_latest_PYRO_latest_EMCEE_latest_TF_latest/lib/python3.8/site-packages/numpy/core/fromnumeric.py in _wrapit(obj, method, *args, **kwds)
     42     except AttributeError:
     43         wrap = None
---> 44     result = getattr(asarray(obj), method)(*args, **kwds)
     45     if wrap:
     46         if not isinstance(result, mu.ndarray):

/usr/local/envs/testenv_3.8_PYSTAN_latest_PYRO_latest_EMCEE_latest_TF_latest/lib/python3.8/site-packages/numpy/core/_methods.py in _clip(a, min, max, out, casting, **kwargs)
    138             um.maximum, a, min, out=out, casting=casting, **kwargs)
    139     else:
--> 140         return _clip_dep_invoke_with_casting(
    141             um.clip, a, min, max, out=out, casting=casting, **kwargs)
    142 

/usr/local/envs/testenv_3.8_PYSTAN_latest_PYRO_latest_EMCEE_latest_TF_latest/lib/python3.8/site-packages/numpy/core/_methods.py in _clip_dep_invoke_with_casting(ufunc, out, casting, *args, **kwargs)
     92     # try to deal with broken casting rules
     93     try:
---> 94         return ufunc(*args, out=out, **kwargs)
     95     except _exceptions._UFuncOutputCastingError as e:
     96         # Numpy 1.17.0, 2019-02-24

ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()

If instead levels is set to an integer specifying the number of contours desired (e.g. levels=3), a different error is thrown.

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-8-eec0f47bee3b> in <module>
      6 
      7 levels = 3 #[0, 0.01, 0.07, 0.13, 0.2]
----> 8 ax = az.plot_kde(
      9     data[:, 0],
     10     data[:, 1],

/opt/arviz/arviz/plots/kdeplot.py in plot_kde(values, values2, cumulative, rug, label, bw, adaptive, circular, quantiles, rotated, contour, hdi_probs, fill_last, figsize, textsize, plot_kwargs, fill_kwargs, rug_kwargs, contour_kwargs, contourf_kwargs, pcolormesh_kwargs, is_circular, ax, legend, backend, backend_kwargs, show, return_glyph, **kwargs)
    341     # TODO: Add backend kwargs
    342     plot = get_plotting_function("plot_kde", "kdeplot", backend)
--> 343     ax = plot(**kde_plot_args)
    344 
    345     return ax

/opt/arviz/arviz/plots/backends/bokeh/kdeplot.py in plot_kde(density, lower, upper, density_q, xmin, xmax, ymin, ymax, gridsize, values, values2, rug, label, quantiles, rotated, contour, fill_last, figsize, textsize, plot_kwargs, fill_kwargs, rug_kwargs, contour_kwargs, contourf_kwargs, pcolormesh_kwargs, is_circular, ax, legend, backend_kwargs, show, return_glyph)
    207                 vertices, _ = contour_generator.create_filled_contour(level, level_upper)
    208                 for seg in vertices:
--> 209                     patch = ax.patch(*seg.T, fill_color=color, **contour_kwargs)
    210                     glyphs.append(patch)
    211 

/usr/local/envs/testenv_3.8_PYSTAN_latest_PYRO_latest_EMCEE_latest_TF_latest/lib/python3.8/site-packages/bokeh/plotting/_decorators.py in wrapped(self, *args, **kwargs)
     79             for arg, param in zip(args, sigparams[1:]):
     80                 kwargs[param.name] = arg
---> 81             return create_renderer(glyphclass, self, **kwargs)
     82 
     83         wrapped.__signature__ = Signature(parameters=sigparams)

/usr/local/envs/testenv_3.8_PYSTAN_latest_PYRO_latest_EMCEE_latest_TF_latest/lib/python3.8/site-packages/bokeh/plotting/_renderer.py in create_renderer(glyphclass, plot, **kwargs)
    115         muted_visuals = None
    116 
--> 117     glyph = make_glyph(glyphclass, kwargs, glyph_visuals)
    118     nonselection_glyph = make_glyph(glyphclass, kwargs, nonselection_visuals)
    119     selection_glyph = make_glyph(glyphclass, kwargs, selection_visuals)

/usr/local/envs/testenv_3.8_PYSTAN_latest_PYRO_latest_EMCEE_latest_TF_latest/lib/python3.8/site-packages/bokeh/plotting/_renderer.py in make_glyph(glyphclass, kws, extra)
    144     kws = kws.copy()
    145     kws.update(extra)
--> 146     return glyphclass(**kws)
    147 
    148 def pop_visuals(glyphclass, props, prefix="", defaults={}, override_defaults={}):

/usr/local/envs/testenv_3.8_PYSTAN_latest_PYRO_latest_EMCEE_latest_TF_latest/lib/python3.8/site-packages/bokeh/model.py in __init__(self, **kwargs)
    234         kwargs.pop("id", None)
    235 
--> 236         super().__init__(**kwargs)
    237         default_theme.apply_to_model(self)
    238 

/usr/local/envs/testenv_3.8_PYSTAN_latest_PYRO_latest_EMCEE_latest_TF_latest/lib/python3.8/site-packages/bokeh/core/has_props.py in __init__(self, **properties)
    267 
    268         for name, value in properties.items():
--> 269             setattr(self, name, value)
    270 
    271         self._initialized = True

/usr/local/envs/testenv_3.8_PYSTAN_latest_PYRO_latest_EMCEE_latest_TF_latest/lib/python3.8/site-packages/bokeh/core/has_props.py in __setattr__(self, name, value)
    303                 matches, text = props, "possible"
    304 
--> 305             raise AttributeError("unexpected attribute '%s' to %s, %s attributes are %s" %
    306                 (name, self.__class__.__name__, text, nice_join(matches)))
    307 

AttributeError: unexpected attribute 'levels' to Patch, possible attributes are fill_alpha, fill_color, hatch_alpha, hatch_color, hatch_extra, hatch_pattern, hatch_scale, hatch_weight, js_event_callbacks, js_property_callbacks, line_alpha, line_cap, line_color, line_dash, line_dash_offset, line_join, line_width, name, subscribed_events, syncable, tags, x or y

Expected behavior In both cases, I’d expect plots to be produced. For comparison, here are the plots produced when backend="matplotlib": for the first instance (where levels = [0, 0.01, 0.07, 0.13, 0.2]) and for the second instance (where levels = 3)

Additional context I used the docker development image, using the ./scripts/container.sh --lab command and a Jupyter Notebook to execute the code. Not sure if it’s relevant but I ran the docker container on Ubuntu 20.04.2 LTS, using Firefox 87.0 to access Jupyter Lab.

[ahartikainen]

Ok, we just probably need to pop levels from contour_dict and use np.asarray

[wm1995]

Oh, sorry, I should have checked back before submitting my PR