Using raw.plot in JupyterLab or Colab

[gurasog]

Is there any recommendation how to use MNE plot() functions such as raw.plot() in JupyterLab and Google Colab?

When i run these functions they aren't interactive. None of %matplotlib %matplotlib qt works in these environment...

So, is there any way to use these graphs interactively in these tools?

Thanks!

[larsoner]

What do you get for mne.sys_info() or mne sys_info on those platforms? Maybe xvfb+Qt5Agg backend would work

[drammock]

I am able to get interactive raw.plot() windows using %matplotlib qt or %matplotlib qt5 in JupyterLab (it pops open a separate window, not an inline figure).

Google Colab is "not officially supported" at the moment, though mainly because of problems with 3D visualization. For 2D visualization it is expected to work the same as Jupyter Notebooks. I tried to get Colab 3D viz working in June, and kept hitting segfaults (there is some discussion of this on the Gitter channel, search for "colab").

[larsoner]

Closing since it seems like it works at least in JupyterLab but let's reopen if it doesn't actually

[larsoner]

@drammock @agramfort @GuillaumeFavelier do one of you want to try what they suggest doing for custom widgets in colab as of a couple of days ago?

googlecolab/colabtools#498 (comment)

I have never used Colab but if it's easy to tell people how to make our IPython widgets work we should do it...

[GuillaumeFavelier]

I can give a try tomorrow. In any case, I'll update #8704 with this. Thanks for the update!

[GuillaumeFavelier]

I managed to reproduce the minimal example using ipympl :

!pip install -q ipympl

%matplotlib widget
from IPython.display import Javascript
import matplotlib.pyplot as plt
import numpy as np

# This enables custom widgets in the current cell's output.
# Currently this must be included in each cell using custom widgets but this
# should be a temporary restriction.
display(Javascript('''
  google.colab.widgets.installCustomManager('http://127.0.0.1:9897/manager.dev.js');
'''))

from mpl_toolkits.mplot3d import axes3d

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

# Grab some test data.
X, Y, Z = axes3d.get_test_data(0.05)

# Plot a basic wireframe.
ax.plot_wireframe(X, Y, Z, rstride=10, cstride=10)

plt.show()

but not the one using ipycanvas (clock.ipynb):

!pip install ipycanvas

from IPython.display import Javascript

display(Javascript('''
  google.colab.widgets.installCustomManager('http://127.0.0.1:9897/manager.dev.js');
'''))

from ipycanvas import Canvas, hold_canvas
import numpy as np

CLOCK_RADIUS = 100

canvas = Canvas(width=CLOCK_RADIUS*2.5, height=CLOCK_RADIUS*2.5)
canvas.translate(CLOCK_RADIUS * 1.2, CLOCK_RADIUS * 1.2)
display(canvas)

def clear_drawing():
    canvas.clear_rect(-CLOCK_RADIUS * 1.2, -CLOCK_RADIUS * 1.2, canvas.width, canvas.height)

def minutes_vec(minutes):
    a = minutes * np.pi/30
    return [np.sin(a), -np.cos(a)]

def draw_dial():
    ht = 10
    mt = 6
    ho = 20
    mo = 10

    canvas.text_align = 'center'
    canvas.text_baseline = 'middle'

    canvas.line_width = 2
    canvas.stroke_circle(0, 0, CLOCK_RADIUS)

    for m in range(60):
        a = m * np.pi/30
        x, y = np.sin(a), -np.cos(a)

        canvas.line_width = 1
        if m % 5 == 0:
            canvas.stroke_line(x * (CLOCK_RADIUS - ht), y * (CLOCK_RADIUS - ht), x * CLOCK_RADIUS, y * CLOCK_RADIUS)
            canvas.font = '12px serif'
            canvas.stroke_text(str(m), x * (CLOCK_RADIUS + mo), y * (CLOCK_RADIUS + mo))
            canvas.font = '16px serif'
            canvas.stroke_text(str(m//5 if m > 0 else 12), x * (CLOCK_RADIUS - ho), y * (CLOCK_RADIUS - ho))
        else:
            canvas.stroke_line(x * (CLOCK_RADIUS - mt), y * (CLOCK_RADIUS - mt), x * CLOCK_RADIUS, y * CLOCK_RADIUS)

def draw_hands(minutes):
    ms = 35
    hs = 50

    hrs = minutes // 60
    mins = minutes % 60

    mv = minutes_vec(mins)
    hv = minutes_vec(hrs * 5 + (mins / 12))

    canvas.line_width = 5
    canvas.stroke_line(0, 0, mv[0] * (CLOCK_RADIUS - ms), mv[1] * (CLOCK_RADIUS - ms))
    canvas.stroke_line(0, 0, hv[0] * (CLOCK_RADIUS - hs), hv[1] * (CLOCK_RADIUS - hs))

def draw_clock(hours, minutes):
    with hold_canvas(canvas):
        clear_drawing()
        draw_dial()
        draw_hands((hours % 12) * 60 + minutes)

import datetime
import ipywidgets as widgets

now = datetime.datetime.now()

hour_text = widgets.IntText(value = now.hour, continuous_update=True, layout={'width': '50px'})
minute_text = widgets.IntText(value = now.minute, continuous_update=True, layout={'width': '50px'})

def on_text_change(change):
    draw_clock(int(hour_text.value), int(minute_text.value))

hour_text.observe(on_text_change, names='value')
minute_text.observe(on_text_change, names='value')

display(widgets.HBox([hour_text, widgets.Label(value=':'), minute_text]))

on_text_change(0)

So the basic example using ipyvtklink did not work either:

!pip install vtk
!pip install ipyvtklink

from IPython.display import Javascript
import vtk
from ipyvtklink.viewer import ViewInteractiveWidget

# This enables custom widgets in the current cell's output.
# Currently this must be included in each cell using custom widgets but this
# should be a temporary restriction.
display(Javascript('''
  google.colab.widgets.installCustomManager('http://127.0.0.1:9897/manager.dev.js');
'''))

# Create some data
cylinder = vtk.vtkCylinderSource()
cylinder.SetResolution(8)
mapper = vtk.vtkPolyDataMapper()
mapper.SetInputConnection(cylinder.GetOutputPort())
actor = vtk.vtkActor()
actor.SetMapper(mapper)

# Set up render window
ren = vtk.vtkRenderer()
ren_win = vtk.vtkRenderWindow()
ren_win.SetOffScreenRendering(1)
ren_win.SetSize(600, 600)
ren_win.AddRenderer(ren)
iren = vtk.vtkRenderWindowInteractor()
iren.SetRenderWindow(ren_win)
style = vtk.vtkInteractorStyleTrackballCamera()
iren.SetInteractorStyle(style)

# Add actor to scene
ren.AddActor(actor)
ren.ResetCamera()

# Display
ViewInteractiveWidget(ren_win)

They both do not give any output.

My bad, I was focused on the 3d plots. I did not try raw.plot 😅

[larsoner]

In the issue linked above they state:

Please feel free to open specific bugs in https://github.com/googlecolab/colab-cdn-widget-manager/issues.

Can you see if this has been reported in https://github.com/googlecolab/colab-cdn-widget-manager/issues and if not, add this information there? Seems like useful debugging for them!

[GuillaumeFavelier]

The related issue is opened in googlecolab/colab-cdn-widget-manager#8 !

[larsoner]

Sounds like they've fixed it: googlecolab/colab-cdn-widget-manager#8 (comment)

[larsoner]

Bumping the milestone of this to the next release, hopefully all (or at least more) bugs will be ironed out by then

[larsoner]

I'll close this as upstream bug(s), but people feel free to comment if it's fixed yet (or not)!