Mailman 3 September 2017 - Matplotlib-users

Windows7: figure window "not responding"
by Strozzi, David J. Oct. 9, 2017

Oct. 9, 2017

This problem has been driving me nuts, posted it elsewhere, so far no one's been able to help. I am running Python 3.6.1 from Anaconda (all packages updated to latest and greatest) on Windows 7. ipython and mpl figure windows work fine from a console, or spyder. The problem is when I try running ipython within emacs using elpy. To back up, the workflow I'm shooting for is using emacs as my IDE, and running python within emacs. I've done this for many years with the Yorick interpreter, and it's quite addictive. I get the same problem whether I use the GNU windows build of emacs, or emacs -nw (terminal, no X) from cygwin (ver 25.2 both cases). elpy seems to be a good Python package for emacs, and nominally supports using ipython. OK, so - I can use ipython + emacs + elpy. The only problem is matplotlib figures. When I do figure(), a figure window appears, but is not fully rendered (e.g. no buttons), and the window title says "Not responding". If I do a plot(), nothing is plotted. Playing with ion() and show() doesn't help. I've posted this on the elpy github, various stackexchange forums, no one has had any ideas yet. Any help is greatly appreciated! Dave

4 11

Problem doing multiple independent plots
by William Ray Wing Sept. 29, 2017

Sept. 29, 2017

Below is a simplified version of a much more elaborate analysis code, but it will illustrate the problem I’m having. What I want to do is repetitively call an analysis function from my main code and plot the results of that analysis (a curve fit, although thats immaterial here) while in that function. Back in the main code, I want to plot the results of all the curve fits on a single plot. They share a common x axis, but appear at different points along it. What seems to be happening is that the gets set in the function, and doesn’t get set back in the main code. Note that there are two versions of the “problem” function, problem and problem_alt. If you change the main code (move the #), you get the plot I want at the end of the main. There must be something I can call or set to recover the settings associated with figure(2), but I can’t seem to figure it out. Any help would be appreciated. Thanks, Bill Wing #! /usr/bin/env python # -*- coding: utf-8 -*- # # A simple skeleton of the program to work out the plotting problem # import numpy as np, matplotlib.pyplot as plt # # skeleton subroutines # def problem(xdata, ydata, i): color_dic = {1: "red", 2: "green", 3: "blue", 4: "cyan"} plt.figure(1) fig1, ax1 = plt.subplots() plt.subplot() plt.plot(xdata, ydata, linestyle = '-', color = color_dic[i]) plt.savefig('Plot for run_num ' + str(i)) return def problem_alt(xdata, ydata, i): return t = np.arange(0.0, 2.0, 0.01) plt.figure(2) fig2, ax2 = plt.subplots() for i in range(0,4): i = i+1 problem(t, np.sin(i*np.pi*3*t), i) problem_alt(t, np.sin(i*np.pi*3*t), i) ax2.set_xlim(xmin = 0.0, xmax = 20.0) plt.subplot() plt.plot((t+i*3), np.sin(i*np.pi*3*(t+i*3))) plt.savefig("Global Plot")

5 7

Window flickering with Tk backend (was: Subplot x-axis shared with other subplots y-axis)
by Juan Nunez-Iglesias Sept. 21, 2017

Sept. 21, 2017

Oops, revisiting this thread now, it looks like I only replied to Ben, instead of the list. Full thread below, but TL;DR: my 3D orthogonal views project: https://github.com/jni/mpl-volume-viewer is working (thanks to all on this list who helped so far!), but has the following problems: - Experiences flickering when panning in the Tk backend - Has no flickering with the MacOS backend, but then the keyboard bindings don’t work at all - Is dog-slow in the nbagg backend. Any suggestions to help with this would be much appreciated! To recap, here are the steps to reproduce: - Download slice_view.py and put it in your PYTHONPATH or current working directory - Download this example dataset: https://www.dropbox.com/s/dj3oarx7bqx4us2/E_z2_512_1um_CONTROL.tif?dl=1 - Run this code: import matplotlib matplotlib.use('TkAgg’) # change this to whatever you want to test from skimage import io import slice_view as sv filename = '/path/to/E_z2_512_1um_CONTROL.tif' image = io.imread(filename) / 4096 v = sv.SliceViewer(image, spacing=[5, 1, 1]) matplotlib.pyplot.show(block=True) Thanks! Juan. ---------- Forwarded message ---------- From: Juan Nunez-Iglesias <jni.soma(a)gmail.com> Date: 11 Jul 2017, 12:06 PM +1000 To: Benjamin Root <ben.v.root(a)gmail.com> Subject: Re: [Matplotlib-users] Subplot x-axis shared with other subplots y-axis > Hi Ben, thanks for the tip! The point of using the ToolManager would be to get cross-backend support, so it might be easier for me to implement the backends I want with if-else statements, or some other kludgey way… > > I just thought I’d update the list with the status of my orthogonal views mini-project, which is backend-dependent: > > - working on Tk except for the flickering > - working on macOS except the keyboard bindings don’t work — they go straight to the Terminal. > - working on Notebook (Nbagg?), except it’s dog-slow. > > If anyone has suggestions for any of these problems, I would be very grateful! > > Juan. > > On 28 Jun 2017, 1:08 AM +1000, Benjamin Root <ben.v.root(a)gmail.com>, wrote: > > re: cross-backend way to add buttons: There is experimental feature called "ToolManager", but I think it is only implemented for a couple of backends currently. You can see an example here: > > > > http://matplotlib.org/examples/user_interfaces/toolmanager.html > > > > > On Fri, Jun 23, 2017 at 12:17 AM, Juan Nunez-Iglesias <jni.soma(a)gmail.com> wrote: > > > > Hey Tom, > > > > > > > > Thanks for that response, it was just what I needed! I had a fun infinite loop where the x-axis was updating the y-axis which was updating the x-axis which... =P But it was easy enough to sort out. =) > > > > > > > > I have a working orthogonal slice viewer here: > > > > https://github.com/jni/mpl-volume-viewer/ > > > > > > > > The problem I have now is that, if I set a mouse-click callback (this line and this line), and then I select any other tool from the toolbar, I get this super-obnoxious flickering, recorded here (macOS 10.12, Tk backend): > > > > https://www.dropbox.com/s/qvfmgp0x36p6rb9/mpl-click-bug.mov?dl=0 > > > > > > > > To reproduce: > > > > > > > > - Download slice_view.py and put it in your PYTHONPATH or current working directory > > > > - Download this example dataset: > > > > https://www.dropbox.com/s/dj3oarx7bqx4us2/E_z2_512_1um_CONTROL.tif?dl=1 > > > > - Run this code: > > > > > > > > import matplotlib > > > > matplotlib.use('TkAgg') > > > > > > > > from skimage import io > > > > import slice_view as sv > > > > > > > > filename = '/path/to/E_z2_512_1um_CONTROL.tif' > > > > image = io.imread(filename) / 4096 > > > > > > > > v = sv.SliceViewer(image, spacing=[5, 1, 1]) > > > > matplotlib.pyplot.show(block=True) > > > > > > > > Questions: > > > > 1) Any ideas about how to fix the flickering? > > > > 2) Is there a cross-backend way to add buttons to the toolbar? All the examples I've found online were backend-specific. (This way I could at least disable the callback at will.) > > > > > > > > Thanks! > > > > > > > > Juan. > > > > > > > > On 16 May 2017, 1:43 AM +1000, Thomas Caswell <tcaswell(a)gmail.com>, wrote: > > > > > There are callbacks that fire when the limits are changed. See > > > > > > > > > > https://matplotlib.org/examples/event_handling/viewlims.html > > > > > > > > > > for an example. > > > > > > > > > > This is probably the best route to go down as you can do other interesting things as the event goes by (like triggering computation, updating a openGL view window, etc) as well as updating the Matplotlib limits. > > > > > > > > > > Tom > > > > > > > > > > > On Mon, May 15, 2017 at 3:29 AM Juan Nunez-Iglesias <jni.soma(a)gmail.com> wrote: > > > > > > > Hello, > > > > > > > > > > > > > > I’m trying to build an orthogonal-views volume viewer in Matplotlib like this one: > > > > > > > https://imagej.nih.gov/ij/docs/guide/images/OrthogonalViews.png > > > > > > > > > > > > > > For this to work, I need to share the y-axis of the YZ (right) view/subplot with the x-axis of the XZ (bottom) view/subplot. Is this possible? > > > > > > > > > > > > > > Juan. > > > > > > > > > > > > > > _______________________________________________ > > > > > > > Matplotlib-users mailing list > > > > > > > Matplotlib-users(a)python.org > > > > > > > https://mail.python.org/mailman/listinfo/matplotlib-users > > > > > > > > _______________________________________________ > > > > Matplotlib-users mailing list > > > > Matplotlib-users(a)python.org > > > > https://mail.python.org/mailman/listinfo/matplotlib-users > > > > > >

1 0

Understanding axes position
by Nunzio Losacco Sept. 20, 2017

Sept. 20, 2017

Hi all, what I’m trying to do here is having third plot with its base aligned with the others and with reduced height (the final aim is custom positioning a colorbar). As you can see I’m getting the wrong position. What am I missing? Thanks for any help NL # -*- coding: utf-8 -*- from __future__ import unicode_literals import numpy as np from scipy.interpolate import griddata import matplotlib import matplotlib.pyplot as plt from matplotlib.patches import Circle from matplotlib.pylab import cm import matplotlib.colors as colors from mpl_toolkits.axes_grid1 import make_axes_locatable, axes_size matplotlib.rcParams.update({'font.size': 8}) fig = plt.figure() fig.set_size_inches(6.3,6.3) ax1 = plt.subplot(111) divider = make_axes_locatable(ax1) ax2 = divider.append_axes('right', size='100%', pad=0.3) axes = [ax1, ax2] ltypes = ['dashed', 'solid'] xi = np.linspace(-18.125, 18.125, 11) yi = np.linspace(0, 28, 9) xv, yv = np.meshgrid(xi, yi) xcOdd = 0.2 zcOdd = 0.725 xcEven = 0.6 zcEven = 0.725 maskRadius = 0.15 for i in range(2): ax = axes[i] ax.set_xlabel('distance [m]') if i == 0: ax.set_ylabel('depth [m]') if i == 1: ax.set_yticklabels([]) ax.invert_yaxis() ax.tick_params(direction='in') ax.set_aspect('equal') odd = Circle((xcOdd, zcOdd), .15, linewidth=1.2, color='k', fill=False) even = Circle((xcEven, zcEven), .15, linewidth=1.2, linestyle=ltypes[i], color='k', fill=False) vmax = 15. vmin = 0. norm = matplotlib.colors.Normalize(vmin,vmax, clip=False) color_map = matplotlib.colors.ListedColormap(plt.cm.Greys(np.linspace(0.25, 1, 5)), "name") ax.add_patch(odd) ax.add_patch(even) pad = 0.03 width = 0.03 pos = ax2.get_position() ax3 = fig.add_axes([pos.xmax + pad, pos.ymin, width, 0.7*(pos.ymax-pos.ymin) ]) plt.savefig('prova-vect-paper-test-2.eps', format='eps')

4 7

matplotlibrc prop_cycle error
by Emlyn Price Sept. 14, 2017

Sept. 14, 2017

Hi, I have been trying to set up a custom color cycle in matplotlib using the matplotlibrc file. I have encountered a problem with the default matplotlibrc file provided with the installation from Arch llinux's package manager and the example provided in the online documentation at: https://matplotlib.org/users/customizing.html When the lines controlling the axes.prop_cycle are uncommented: axes.prop_cycle : cycler('color', ['1f77b4', 'ff7f0e', '2ca02c', 'd62728', '9467bd', '8c564b', 'e377c2', '7f7f7f', 'bcbd22', '17becf']) the following message is generated when I import matplotlib: /usr/lib/python3.6/site-packages/matplotlib/__init__.py:1078: UserWarning: Illegal line #338 " ['1f77b4', 'ff7f0e', '2ca02c', 'd62728', " in file "/home/ejp/mpl_test/matplotlibrc" warnings.warn('Illegal %s' % error_details) /usr/lib/python3.6/site-packages/matplotlib/__init__.py:1078: UserWarning: Illegal line #339 " '9467bd', '8c564b', 'e377c2', '7f7f7f', " in file "/home/ejp/mpl_test/matplotlibrc" warnings.warn('Illegal %s' % error_details) /usr/lib/python3.6/site-packages/matplotlib/__init__.py:1078: UserWarning: Illegal line #340 " 'bcbd22', '17becf']) " in file "/home/ejp/mpl_test/matplotlibrc" warnings.warn('Illegal %s' % error_details) /usr/lib/python3.6/site-packages/matplotlib/__init__.py:1120: UserWarning: Bad val "cycler('color'," on line #337 "axes.prop_cycle : cycler('color', " in file "/home/ejp/mpl_test/matplotlibrc" Key axes.prop_cycle: 'cycler('color',' is not a valid cycler construction: unexpected EOF while parsing (<string>, line 1) (val, error_details, msg)) The same error is generated if I import cycler using import cycler or from cycler import cycler I am using matplotlib 2.0.2 in python version 3.6.2 installed from the Arch Linux repositories. The ouput of uname -a is :Linux tinker 4.12.12-1-ARCH #1 SMP PREEMPT Sun Sep 10 09:41:14 CEST 2017 x86_64 GNU/Linux To reproduce: Copy the default matplotlibrc Uncomment the lines containing axes.prop_cycle : cycler('color', ['1f77b4', 'ff7f0e', '2ca02c', 'd62728', '9467bd', '8c564b', 'e377c2', '7f7f7f', 'bcbd22', '17becf']) Run python in the directory containing the matplotlibrc file Import matplotlib Am I missing something obvious? Thanks, Emlyn

3 2

Ticks at fixed locations with uneven interval
by erzdemir Sept. 13, 2017

Sept. 13, 2017

I would like to keep ticks at locations evenly spaced but keep their values uneven. Is it possible to do it in matplotlib. -- Sent from: http://matplotlib.1069221.n5.nabble.com/matplotlib-users-f3.html

3 2

Saving interactive plot
by cdpnepal Sept. 8, 2017

Sept. 8, 2017

Is there a way to save interactive plot in matplotlib -- Sent from: http://matplotlib.1069221.n5.nabble.com/matplotlib-users-f3.html

3 2

selecting points inside plot
by Jean-Philippe Grivet Sept. 8, 2017

Sept. 8, 2017

Hello, I use Win7 and the Anaconde suite of Python programs. I am trying to write a piece of software to demonstrate the computation of electric potentials. The software should ; - allow a user to select (with a mouse click) a point inside a blank rectangle, - record the x and y coordinates of that point - draw a dot at that location - ask the user for the value of the charge located there and record that - repeat a few times - compute the electric potential and draw a contour plot of that function. There are several pieces of code available showing how to print the x,y coor- dinates of a mouse click but I can't devise a way to retrieve thes values for turther computations nor can I dra a dot at the corresponding location or insert a dialog to get the charge values. I would be grateful for any suggestion. JP Grivet --- L'absence de virus dans ce courrier électronique a été vérifiée par le logiciel antivirus Avast. https://www.avast.com/antivirus

5 9

Circular Arrow Closed Path
by Amit Yaron Sept. 7, 2017

Sept. 7, 2017

Hi, Does this shape already exist in Matplotlib modules? I've attached an example prepared with LibreOffice Draw.

2 2

[ANN]: Matplotlib 2.1.0rc1
by Thomas Caswell Sept. 2, 2017

Sept. 2, 2017

Folks, We are happy to announce 2.1.0rc1! Wheels and and source on pypi and as a conda package via conda-forge. Please try it out and report any issues! See http://matplotlib.org/devdocs/users/whats_new.html for a draft of the release notes. This represents ~2 years of work and includes many new features and bug fixes. Please checkout our updated documentation ( http://matplotlib.org/devdocs/index.html), particularly the examples ( http://matplotlib.org/devdocs/gallery/index.html ) and tutorials ( http://matplotlib.org/devdocs/tutorials/index.html) . Re-organizing, updating, and migrating the examples to sphinx-gallery was a huge effort led by Nelle Varoquaux and Chris Holdgraf. A big thanks to the many (~280) people who contributed to this release. Installation instructions (via pip into existing venv): pip install --upgrade --pre matplotlib or (via conda into a new venv) conda create -n mpl_rc matplotlib ipython -c conda-forge/label/rc -c conda-forge or(via conda into an existing venv) conda install -c conda-forge/label/rc matplotlib -c conda-forge Tom

3 4