[Bug]: AttributeError: 'Arrow3D' object has no attribute 'do_3d_projection' · Issue #21688 · matplotlib/matplotlib · GitHub

Bug summary

New (appears in matplotlib=3.5) an AttributeError is thrown stating 'Arrow3D' object has no attribute 'do_3d_projection'.

The error appears when running, e.g., the nice example https://stackoverflow.com/a/29188796 by @tacaswell . Below I extracted a MWE from the stackoverflow code.

Code for reproduction

# MWE extracted from https://stackoverflow.com/a/29188796 by @tacaswell 

import numpy as np
from matplotlib import pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
from matplotlib.patches import FancyArrowPatch
from mpl_toolkits.mplot3d import proj3d

class Arrow3D(FancyArrowPatch):
    def __init__(self, xs, ys, zs, *args, **kwargs):
        FancyArrowPatch.__init__(self, (0,0), (0,0), *args, **kwargs)
        self._verts3d = xs, ys, zs

    def draw(self, renderer):
        xs3d, ys3d, zs3d = self._verts3d
        xs, ys, zs = proj3d.proj_transform(xs3d, ys3d, zs3d, renderer.M)
        self.set_positions((xs[0],ys[0]),(xs[1],ys[1]))
        FancyArrowPatch.draw(self, renderer)

fig = plt.figure()
ax = fig.add_subplot(111, projection='3d')
arrow_prop_dict = dict(mutation_scale=20, arrowstyle='-|>', color='k', shrinkA=0, shrinkB=0)
a = Arrow3D([0, 10], [0, 0], [0, 0], **arrow_prop_dict)
ax.add_artist(a)

Actual outcome

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
~/anaconda3/lib/python3.8/site-packages/IPython/core/formatters.py in __call__(self, obj)
    339                 pass
    340             else:
--> 341                 return printer(obj)
    342             # Finally look for special method names
    343             method = get_real_method(obj, self.print_method)

~/anaconda3/lib/python3.8/site-packages/IPython/core/pylabtools.py in <lambda>(fig)
    248 
    249     if 'png' in formats:
--> 250         png_formatter.for_type(Figure, lambda fig: print_figure(fig, 'png', **kwargs))
    251     if 'retina' in formats or 'png2x' in formats:
    252         png_formatter.for_type(Figure, lambda fig: retina_figure(fig, **kwargs))

~/anaconda3/lib/python3.8/site-packages/IPython/core/pylabtools.py in print_figure(fig, fmt, bbox_inches, **kwargs)
    132         FigureCanvasBase(fig)
    133 
--> 134     fig.canvas.print_figure(bytes_io, **kw)
    135     data = bytes_io.getvalue()
    136     if fmt == 'svg':

~/anaconda3/lib/python3.8/site-packages/matplotlib/backend_bases.py in print_figure(self, filename, dpi, facecolor, edgecolor, orientation, format, bbox_inches, pad_inches, bbox_extra_artists, backend, **kwargs)
   2288                 )
   2289                 with getattr(renderer, "_draw_disabled", nullcontext)():
-> 2290                     self.figure.draw(renderer)
   2291 
   2292             if bbox_inches:

~/anaconda3/lib/python3.8/site-packages/matplotlib/artist.py in draw_wrapper(artist, renderer, *args, **kwargs)
     71     @wraps(draw)
     72     def draw_wrapper(artist, renderer, *args, **kwargs):
---> 73         result = draw(artist, renderer, *args, **kwargs)
     74         if renderer._rasterizing:
     75             renderer.stop_rasterizing()

~/anaconda3/lib/python3.8/site-packages/matplotlib/artist.py in draw_wrapper(artist, renderer)
     48                 renderer.start_filter()
     49 
---> 50             return draw(artist, renderer)
     51         finally:
     52             if artist.get_agg_filter() is not None:

~/anaconda3/lib/python3.8/site-packages/matplotlib/figure.py in draw(self, renderer)
   2801 
   2802             self.patch.draw(renderer)
-> 2803             mimage._draw_list_compositing_images(
   2804                 renderer, self, artists, self.suppressComposite)
   2805 

~/anaconda3/lib/python3.8/site-packages/matplotlib/image.py in _draw_list_compositing_images(renderer, parent, artists, suppress_composite)
    130     if not_composite or not has_images:
    131         for a in artists:
--> 132             a.draw(renderer)
    133     else:
    134         # Composite any adjacent images together

~/anaconda3/lib/python3.8/site-packages/matplotlib/artist.py in draw_wrapper(artist, renderer)
     48                 renderer.start_filter()
     49 
---> 50             return draw(artist, renderer)
     51         finally:
     52             if artist.get_agg_filter() is not None:

~/anaconda3/lib/python3.8/site-packages/mpl_toolkits/mplot3d/axes3d.py in draw(self, renderer)
    445                                     for axis in self._get_axis_list()) + 1
    446                 collection_zorder = patch_zorder = zorder_offset
--> 447                 for artist in sorted(collections_and_patches,
    448                                      key=do_3d_projection,
    449                                      reverse=True):

~/anaconda3/lib/python3.8/site-packages/mpl_toolkits/mplot3d/axes3d.py in do_3d_projection(artist)
    434                     "do_3d_projection() was deprecated in Matplotlib "
    435                     "%(since)s and will be removed %(removal)s.")
--> 436                 return artist.do_3d_projection(renderer)
    437 
    438             collections_and_patches = (

AttributeError: 'Arrow3D' object has no attribute 'do_3d_projection'

[@tacaswell edited to add markup]

Expected outcome

Should draw the fancy arrow.

Additional information

This worked before matplotlib 3.5; error appears in different Python versions and different backends.

Operating system

Linux

Matplotlib Version

3.5.0

Matplotlib Backend

module://ipykernel.pylab.backend_inline

Python version

No response

Jupyter version

No response

Installation

conda

Do you know what version of Matplotlib this last worked with? Thanks!

It starts to fail with v3.5.0. v3.4.3 is still fine.

I think this is another bit of fallout from merging the internal representation of axes children from any lists to 1 list. Previously, the Arrow3D here would be in a list that was did not participate in the z-order sorting, and hence that it was missing one of the methods from the 3D api was missed. However, now that we have them merged, we reconstruct the old lists (which were nominally sorted by type) by picking out all of the children of the right type.

The fix here is to correct Arrow3D to have the missing method:

import numpy as np
from matplotlib import pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
from matplotlib.patches import FancyArrowPatch
from mpl_toolkits.mplot3d import proj3d

class Arrow3D(FancyArrowPatch):
    def __init__(self, xs, ys, zs, *args, **kwargs):
        super().__init__((0,0), (0,0), *args, **kwargs)
        self._verts3d = xs, ys, zs

    def do_3d_projection(self, renderer=None):
        xs3d, ys3d, zs3d = self._verts3d
        xs, ys, zs = proj3d.proj_transform(xs3d, ys3d, zs3d, self.axes.M)
        self.set_positions((xs[0],ys[0]),(xs[1],ys[1]))

        return np.min(zs)

fig = plt.figure()
ax = fig.add_subplot(111, projection='3d')
arrow_prop_dict = dict(mutation_scale=20, arrowstyle='-|>', color='k', shrinkA=0, shrinkB=0)
a = Arrow3D([0, 10], [0, 0], [0, 0], **arrow_prop_dict)
ax.add_artist(a)

plt.show()

This works fine, thanks @tacaswell !

    def do_3d_projection(self, render=None):

That should be renderer, not render, though I don't believe anything calls it with the name explicitly.

Add deprecated 3d projection method back to _Arrow3D: matplotlib/matp…

Fix dxtbx.plot_detector_models for matplotlib 3.5

upgrade to last mathplotlib versions

Create Listing_3-1_new.py

Exact Code .. here

%pylab inline

from matplotlib import pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
from mpl_toolkits.mplot3d import proj3d
from matplotlib.patches import FancyArrowPatch
%matplotlib inline

class Arrow3D(FancyArrowPatch):
def init(self, xs, ys, zs, *args, **kwargs):
FancyArrowPatch.init(self, (0,0), (0,0), *args, **kwargs)
self._verts3d = xs, ys, zs

def  do_3d_projection(self, renderer):
    xs3d, ys3d, zs3d = self._verts3d
    xs, ys, zs = proj3d.proj_transform(xs3d, ys3d, zs3d, self.axes.M)
    self.set_positions((xs[0],ys[0]),(xs[1],ys[1]))
    FancyArrowPatch.draw(self, renderer)

    return np.min(zs)

fig = plt.figure(figsize=(7,7))
ax = fig.add_subplot(111, projection='3d')

ax.plot(df['feature1'], df['feature2'], df['feature3'], 'o', markersize=8, color='blue', alpha=0.2)
ax.plot([df['feature1'].mean()], [df['feature2'].mean()], [df['feature3'].mean()], 'o', markersize=10, color='red', alpha=0.5)
for v in eigen_vectors.T:
a = Arrow3D([df['feature1'].mean(), v[0]], [df['feature2'].mean(), v[1]], [df['feature3'].mean(), v[2]], mutation_scale=20, lw=3, arrowstyle="-|>", color="r")
ax.add_artist(a)
ax.set_xlabel('x_values')
ax.set_ylabel('y_values')
ax.set_zlabel('z_values')

plt.title('Eigenvectors')

plt.show()

@Siddhartha082 I do not understand what you are trying to say.

Thanks you @tacaswell !

I think this is another bit of fallout from merging the internal representation of axes children from any lists to 1 list. Previously, the Arrow3D here would be in a list that was did not participate in the z-order sorting, and hence that it was missing one of the methods from the 3D api was missed. However, now that we have them merged, we reconstruct the old lists (which were nominally sorted by type) by picking out all of the children of the right type.

The fix here is to correct Arrow3D to have the missing method:
import numpy as np
from matplotlib import pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
from matplotlib.patches import FancyArrowPatch
from mpl_toolkits.mplot3d import proj3d

class Arrow3D(FancyArrowPatch):
    def __init__(self, xs, ys, zs, *args, **kwargs):
        super().__init__((0,0), (0,0), *args, **kwargs)
        self._verts3d = xs, ys, zs

    def do_3d_projection(self, renderer=None):
        xs3d, ys3d, zs3d = self._verts3d
        xs, ys, zs = proj3d.proj_transform(xs3d, ys3d, zs3d, self.axes.M)
        self.set_positions((xs[0],ys[0]),(xs[1],ys[1]))

        return np.min(zs)

fig = plt.figure()
ax = fig.add_subplot(111, projection='3d')
arrow_prop_dict = dict(mutation_scale=20, arrowstyle='-|>', color='k', shrinkA=0, shrinkB=0)
a = Arrow3D([0, 10], [0, 0], [0, 0], **arrow_prop_dict)
ax.add_artist(a)

plt.show()

NB,NB,NB

Can this class just be added to the library?

Fix using

fix object has no attribute 'do_3d_projection'.

[Bug]: AttributeError: 'Arrow3D' object has no attribute 'do_3d_projection' · Issue #21688 · matplotlib/matplotlib · GitHub

Repository navigation

[Bug]: AttributeError: 'Arrow3D' object has no attribute 'do_3d_projection' #21688

Bug summary

Code for reproduction

Actual outcome

Expected outcome

Additional information

Operating system

Matplotlib Version

Matplotlib Backend

Python version

Jupyter version

Installation

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

[Bug]: AttributeError: 'Arrow3D' object has no attribute 'do_3d_projection' · Issue #21688 · matplotlib/matplotlib · GitHub

Repository navigation

[Bug]: AttributeError: 'Arrow3D' object has no attribute 'do_3d_projection' #21688

Description

Bug summary

Code for reproduction

Actual outcome

Expected outcome

Additional information

Operating system

Matplotlib Version

Matplotlib Backend

Python version

Jupyter version

Installation

Activity

jklymak commented on Nov 20, 2021

prisae commented on Nov 20, 2021

tacaswell commented on Nov 21, 2021

prisae commented on Nov 22, 2021

QuLogic commented on Nov 22, 2021

Siddhartha082 commented on Aug 26, 2022

tacaswell commented on Aug 26, 2022

aimlwithgautam commented on Nov 8, 2022

Neombpure commented on Mar 5, 2023

gabemorris12 commented on Mar 22, 2024

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions