Mailman 3 September 2019 - Matplotlib-users

Unusual layout (GridSpec?)
by Bruno Pagani Sept. 26, 2019

Sept. 26, 2019

Hi there, I’m trying to combine three graphs into one, but with some specific constraints on the layout. The three graphs are having the same x and y variables, but not over the same span. Each graph size should reflects that (e.g. if a change of 1 on the x axis is 1 cm on one graph, so should it be for the other ones). Also, I’d like the y label and ticks labels to be shared amongst the three graphs. Same goes for the colorbar (those are scatter plots) and the legend. Attached are the three figures (.pdf), and what the combined version should look like (target.png, but here it was obtained by removing parts by hand and stretching everything to fit). Below is some minimal code to produce three similarly looking graphs, if someone has an idea (GridSpec?) for this, I should be able to adapt it to my actual code: import matplotlib.pyplot as plt import numpy as np plt.scatter(np.repeat(np.linspace(0,3.5,15)[:,None],11,axis=1),np.repeat(np.linspace(0,0.5,11)[None,:],15,axis=0),c=np.random.random((15,11)),label="Some label") plt.scatter(np.repeat(np.linspace(0,6,25)[:,None],11,axis=1),np.repeat(np.linspace(0,0.5,11)[None,:],25,axis=0),c=np.random.random((25,11)),marker='*',label="Some other label") plt.scatter(np.repeat(np.linspace(0,11,23)[:,None],11,axis=1),np.repeat(np.linspace(0,0.5,11)[None,:],23,axis=0),c=np.random.random((23,11)),marker='D',label="Yet another label") Regards, Bruno

3 4

install under Python 3.8
by Alan Isaac Sept. 24, 2019

Sept. 24, 2019

Should it be possible to install matplotlib with pip under Python 3.8b4? It failed for me, and it looks like the installer ran into problems when calling cl.exe: "failed with exit status 2" Thank you, Alan Isaac

3 4

pyplot.contour() and filtered levels
by Olivier B. Sept. 24, 2019

Sept. 24, 2019

At the switch to MPL 2.2 (from 1.5 i think), my calls to pyplot.clabel() started throwing ValueError. This turned out to be because the pyplot.contour() returned a ContourSet which had filtered out the levels that were not displayed in the area of the data i am trying to plot. As i build my list of level labels for the list of levels that i pass to contour(), that list could contain levels that had been filtered out by contour(), which throws a ValueError in clabel(). so i also filtered my list of levels to label based on the levels kept by the CountourSet. Now, i get the impression that there is a similar issue for the optional lists of line width, style, and color that we pass to pyplot.contour. If i move/zoom in a specific area of my data to draw isolines on it, with the same python code everytime, that uses the same list of levels based on the data of the whole area, and not levels based on the data on the drawn area, depending on the area/zoom, the lines styles seem to be shifted. As if levels were filtered out, but not the related items of the other arrays, so styles apply to a different level. Does someone have a hint about what's actually going on and how i might work around it, or, if i'm lucky, if a version more recent than 2.2.3 fixes this issue?

2 2

ANN: mplcairo 0.2 release
by Antony Lee Sept. 23, 2019

Sept. 23, 2019

Dear all, I am pleased to announce the release of mplcairo 0.2. mplcairo is a Matplotlib backend based on the well-known cairo library, supporting output to both raster (including interactively) and vector formats. In other words, it provides the functionality of Matplotlib's {,qt5,gtk3,wx,tk,macos}{agg,cairo}, pdf, ps, and svg backends. Per Matplotlib's standard API, the backend can be selected by calling matplotlib.use("module://mplcairo.qt") or setting your MPLBACKEND environment variable to `module://mplcairo.qt` for Qt5, and similarly for other toolkits. mplcairo 0.2 adds support for cairo 1.17.2's high-precision floating point surfaces, simplifies the use of custom compositing operators (see `examples/operators.py`), a few other features listed in the changelog, as well as the usual bugfixes over 0.1. Enjoy, Antony Lee

1 0

New Matplotlib’s toolmanager Works Well ! - How can we attach a callback and use it with tkinter ?
by Carlos Baumann Sept. 19, 2019

Sept. 19, 2019

Hello Matplotlib Users ! The new ToolManager works well, it is very easy to add new ToolButtons ! See snapshot and associated code. Questions: 1. Can we use this with Tkinter ? If yes can you send or point me to one example ? (I know how to use a NavigationToolbar2Tk() with tkinter, but not a ToolbarTk() ) 1. How do we attach a callback function to the new button ? Thanks! [cid:image001.png@01D56EC3.1069B9A0] # From the bottom of: # https://stackoverflow.com/questions/20711148/ignore-matplotlib-cursor-widge… # This works well, and the Toggle Buttons remain Toggled, depressed. import matplotlib.pyplot as plt # This is important plt.rcParams['toolbar'] = 'toolmanager' # # https://fossies.org/linux/matplotlib/lib/matplotlib/backends/_backend_tk.py # 516 if matplotlib.rcParams['toolbar'] == 'toolbar2': # 517 toolbar = NavigationToolbar2Tk(self.canvas, self.window) # 518 elif matplotlib.rcParams['toolbar'] == 'toolmanager': # 519 toolbar = ToolbarTk(self.toolmanager, self.window) fig, ax = plt.subplots() # print(fig.canvas.manager.toolbar._groups) # {'navigation': <tkinter.Frame object .!toolbartk.!frame>, 'zoompan': <tkinter.Frame object .!toolbartk.!frame3>, # 'io': <tkinter.Frame object .!toolbartk.!frame5>} # 807 def add_toolitem(self, name, group, position, image_file, description, toggle): fig.canvas.manager.toolbar.add_toolitem('newCB', 'newg', 0, None, 'new check-button', True) # creates new group and check-button !!! fig.canvas.manager.toolbar.add_toolitem('newB', 'newg', 0, None, 'new button', False) # creates new group and button !!! fig.canvas.manager.toolbar.add_toolitem('newCB', 'navigation', 0, None, 'new check-button', True) # creates button at the end of the first group def on_click(evt): state = fig.canvas.manager.toolbar.toolmanager.active_toggle["default"] if state is None: print("no tool selected") else: print(f"{state} selected") # zoom selected # pan selected # no tool selected cid = fig.canvas.mpl_connect('button_press_event', on_click) plt.show() Schlumberger-Private

1 0

Re: [Matplotlib-users] removing paths inside polygon
by Carl Sept. 18, 2019

Sept. 18, 2019

This is an example that I reviewed in my effort to draw a polygon with certain zones masked, or clipped. I have tried to find a way to Mask polygonal areas before filling/drawing, but so far I have not found what I really need. I need to be able to fill plot areas without touching certain polygonal areas that may be totally or partially where the polygon will be filled. I cannot fill with white the polygonal areas after rendering because I would be deleting information (e.g. colors and zorder) in those areas. Can I copy the information in those polygonal areas, and then copy them back after filling/drawing what I needed ? Here are the bugs issues that I found: This example was found at: http://matplotlib.1069221.n5.nabble.com/removing-paths-inside-polygon-td406… While doing this, I stumbled on a bug and a couple of issues: 1) When .set_clip_path() is used, the "Zoom to rectangle" and "Pan" functions can move draw the plots outside the subplot. See the attached 3 snapshots. 2) “Pick events” are generated even for the "clipped" areas of the star. It would be nice if this was not so. import matplotlib.pyplot as plt import matplotlib.path as mpath import matplotlib.collections as mcol import numpy as np import copy pick = 0 def on_pick(event): global pick # however pick events are NOT generated while the Zoom or Translate options are on ! # https://matplotlib.org/3.1.1/gallery/event_handling/pick_event_demo2.html pick += 1 #print("on_pick ",pick,": ", str(event.artist), " event=", str(event)) print("on_pick ",pick,": zorder:", event.artist.get_zorder()) exterior = mpath.Path.unit_rectangle() exterior = mpath.Path(copy.deepcopy(exterior.vertices), copy.deepcopy(exterior.codes[:])) exterior.vertices *= 4 exterior.vertices -= 2 #interior = mpath.Path.unit_circle() interior = mpath.Path.unit_rectangle() interior = mpath.Path(copy.deepcopy(interior.vertices), copy.deepcopy(interior.codes[:])) interior.vertices = interior.vertices[::-1] # reverse the list vertices = np.concatenate([exterior.vertices, interior.vertices]) codes = np.concatenate([exterior.codes, interior.codes]) interior = mpath.Path.unit_rectangle() interior = mpath.Path(copy.deepcopy(interior.vertices), copy.deepcopy(interior.codes[:])) interior.vertices = interior.vertices[::-1] interior.vertices -= 0.6 vertices = np.concatenate([vertices, interior.vertices]) codes = np.concatenate([codes, interior.codes]) clip_path = mpath.Path(vertices, codes) star = mpath.Path.unit_regular_star(6) star = mpath.Path(copy.deepcopy(star.vertices), copy.deepcopy(star.codes[:])) star.vertices *= 3.5 figu, ax = plt.subplots(figsize=(6,6), dpi=100) cbid = figu.canvas.mpl_connect("pick_event", on_pick) ax = plt.subplot(221) col = mcol.PathCollection([clip_path], facecolor='green') ax.add_collection(col) ax.set_title('Clip path') ax.autoscale() ax = plt.subplot(222) col = mcol.PathCollection([star], facecolor='red') ax.add_collection(col) ax.set_title('Target polygon') ax.autoscale() # BUG: "zoom to rectangle" on this subplot shows part of the star outside this subplot ? # BUG: "Pan" on the clipped star can be translated to anywhere over the other subplots ? # "Zoom to rect" and "Pan" work as expected on the two top subplots. # This is an issue associated with set_clip_path(), when the line below is commented, Zoom / Pan behave as expected. # NOTE: pick events are generated even for the "clipped" areas of the star ! It would be nice if this was not so. ax = plt.subplot(2,2,4) col = mcol.PathCollection([star]) col.set_clip_path(clip_path, ax.transData) col.set(zorder=9, facecolor="blue", picker=True) ax.add_collection(col) ax.set_title('Target polygon clipped by clip_path') ax.autoscale() plt.tight_layout() plt.show() <http://matplotlib.1069221.n5.nabble.com/file/t5261/F1.gif> <http://matplotlib.1069221.n5.nabble.com/file/t5261/F2.gif> <http://matplotlib.1069221.n5.nabble.com/file/t5261/F3.gif> -- Sent from: http://matplotlib.1069221.n5.nabble.com/matplotlib-users-f3.html

1 0

Issues Removing Paths from a filled polygon
by Carlos Baumann Sept. 18, 2019

Sept. 18, 2019

This is an example that I reviewed in my effort to draw a polygon with certain zones masked, or clipped. I have tried to find a way to Mask polygonal areas before filling/drawing, but so far I have not found what I really need. I need to be able to fill plot areas without touching certain polygonal areas that may be totally or partially where the polygon will be filled. I cannot fill with white the polygonal areas after rendering because I would be deleting information (e.g. colors and zorder) in those areas. Can I copy the information in those polygonal areas, and then copy them back after filling/drawing what I needed ? Here are the bugs issues that I found: This example was found at: http://matplotlib.1069221.n5.nabble.com/removing-paths-inside-polygon-td406… While doing this, I stumbled on a bug and a couple of issues: 1. When .set_clip_path() is used, the "Zoom to rectangle" and "Pan" functions can move draw the plots outside the subplot. See the attached 3 snapshots. 1. "Pick events" are generated even for the "clipped" areas of the star. It would be nice if this was not so. import matplotlib.pyplot as plt import matplotlib.path as mpath import matplotlib.collections as mcol import numpy as np import copy pick = 0 def on_pick(event): global pick # however pick events are NOT generated while the Zoom or Translate options are on ! # https://matplotlib.org/3.1.1/gallery/event_handling/pick_event_demo2.html pick += 1 #print("on_pick ",pick,": ", str(event.artist), " event=", str(event)) print("on_pick ",pick,": zorder:", event.artist.get_zorder()) exterior = mpath.Path.unit_rectangle() exterior = mpath.Path(copy.deepcopy(exterior.vertices), copy.deepcopy(exterior.codes[:])) exterior.vertices *= 4 exterior.vertices -= 2 #interior = mpath.Path.unit_circle() interior = mpath.Path.unit_rectangle() interior = mpath.Path(copy.deepcopy(interior.vertices), copy.deepcopy(interior.codes[:])) interior.vertices = interior.vertices[::-1] # reverse the list vertices = np.concatenate([exterior.vertices, interior.vertices]) codes = np.concatenate([exterior.codes, interior.codes]) interior = mpath.Path.unit_rectangle() interior = mpath.Path(copy.deepcopy(interior.vertices), copy.deepcopy(interior.codes[:])) interior.vertices = interior.vertices[::-1] interior.vertices -= 0.6 vertices = np.concatenate([vertices, interior.vertices]) codes = np.concatenate([codes, interior.codes]) clip_path = mpath.Path(vertices, codes) star = mpath.Path.unit_regular_star(6) star = mpath.Path(copy.deepcopy(star.vertices), copy.deepcopy(star.codes[:])) star.vertices *= 3.5 figu, ax = plt.subplots(figsize=(6,6), dpi=100) cbid = figu.canvas.mpl_connect("pick_event", on_pick) ax = plt.subplot(221) col = mcol.PathCollection([clip_path], facecolor='green') ax.add_collection(col) ax.set_title('Clip path') ax.autoscale() ax = plt.subplot(222) col = mcol.PathCollection([star], facecolor='red') ax.add_collection(col) ax.set_title('Target polygon') ax.autoscale() # BUG: "zoom to rectangle" on this subplot shows part of the star outside this subplot ? # BUG: "Pan" on the clipped star can be translated to anywhere over the other subplots ? # "Zoom to rect" and "Pan" work as expected on the two top subplots. # This is an issue associated with set_clip_path(), when the line below is commented, Zoom / Pan behave as expected. # NOTE: pick events are generated even for the "clipped" areas of the star ! It would be nice if this was not so. ax = plt.subplot(2,2,4) col = mcol.PathCollection([star]) col.set_clip_path(clip_path, ax.transData) col.set(zorder=9, facecolor="blue", picker=True) ax.add_collection(col) ax.set_title('Target polygon clipped by clip_path') ax.autoscale() plt.tight_layout() plt.show() [cid:image001.png@01D56DA8.4FE293C0] [cid:image002.png@01D56DA8.4FE293C0] [cid:image003.png@01D56DA8.4FE293C0] Schlumberger-Private

1 0

Does xticks start marking from 0 only?
by Sreyan Chakravarty Sept. 7, 2019

Sept. 7, 2019

Given the code: plt.axis([-5, 105, 0, 5]) # x-axis from -5 to 105, # y-axis from 0 to 5 plt.xticks([10 * i for i in range(11)]) I see xticks only marks from 0, where as the axis starts from -5. So does xticks only start marking from 0 even though the x-axis starts from -5 ? Full Code: from matplotlib import pyplot as plt from collections import Counter grades = [83, 95, 91, 87, 70, 0, 85, 82, 100, 67, 73, 77, 0] # Bucket grades by decile, but put 100 in with the 90s histogram = Counter(min(grade // 10 * 10, 90) for grade in grades) plt.bar([x + 5 for x in histogram.keys()], # Shift bars right by 5 histogram.values(), # Give each bar its correct height 10, # Give each bar a width of 10 edgecolor=(0, 0, 0)) # Black edges for each bar plt.axis([-5, 105, 0, 5]) # x-axis from -5 to 105, # y-axis from 0 to 5 #xticks takes the highest value in the list given and then plt.xticks([10 * i for i in range(11)]) # x-axis labels at 0, 10, ..., 100 plt.xlabel("Decile") plt.ylabel("# of Students") plt.title("Distribution of Exam 1 Grades") plt.show() -- Regards, Sreyan Chakravarty

2 1

How does the x coordinates work for plt.bar()?
by Sreyan Chakravarty Sept. 6, 2019

Sept. 6, 2019

I am unable to understand how the plt.bar() function actually works. Take the following code for example: from collections import Countergrades = [83, 95, 91, 87, 70, 0, 85, 82, 100, 67, 73, 77, 0] # Bucket grades by decile, but put 100 in with the 90shistogram = Counter(min(grade // 10 * 10, 90) for grade in grades) plt.bar([x + 5 for x in histogram.keys()], # Shift bars right by 5 histogram.values(), # Give each bar its correct height 10, # Give each bar a width of 10 edgecolor=(0, 0, 0)) # Black edges for each bar What does it mean by shift bars by 5 in the plt.bar() call? Does it mean by 5 pixels? 5%? 5% of what? These measurements make no sense. Can someone explain to me how this works? -- Regards, Sreyan Chakravarty

2 1

Changing the colorbar / Absolute marker size
by Guy Léonis Sept. 3, 2019

Sept. 3, 2019

Thank you to all, and my deepest apologies to have not found related topics in archive and through googling. I have an Earth map with markers on it. Each marker has 4 attributes: longitude, latitude, temperature and size. In the example, positions are random. [cid:image001.png@01D561B5.0BD968B0] The colorbar range is set to temperature range. When I zoom in the map, I would like to: - change the displayed colorbar so that its range is set to the temperature range of the still visible markers, and change accordingly the colour of the still visible markers - (not yet searched for) have an absolute size for the marker (i.e. diameter is absolute on Earth) Despite search in documentation and examples, I am unable to find a programming path. Any suggestion would be welcomed. Current (dirty) test code of my best solution to first point is appended to this email. Again, thank you. Best regards, Guy [http://www.spacebel.be/wp-content/uploads/2011/06/image-sign-sbp.jpg] Guy Léonis Project Manager Ildefons Vandammestraat, 5-7 Bat B 1st floor - Hoeilaart Office Center - B-1560 HOEILAART Tel: +32 (0) 2 658 20 11 - Fax: +32 (0) 2 658 20 90 www.spacebel.be<http://www.spacebel.be/> # Import python standard libraries import os import sys # Import specific libraries import netCDF4 import matplotlib #matplotlib.use('Qt4Agg') import matplotlib.pyplot as plt import numpy as np pat_version = "2" window_id = 1 channel = 1 sample = 1 extra = "user string" window_title = "MWS - PAT - V%s - %d - Map of channel %02d / sample %02d - %s " \ % (pat_version, window_id, channel, sample, extra) lon = np.random.random(50) * 360 - 180 lat = np.random.random(50) * 180 - 90 bt = np.random.random(50) * 50 + 250 radius = np.random.random(50) (fig, ax) = plt.subplots() sc = None cm = plt.get_cmap("rainbow") def on_draw(event): global sc print("\n\nDraw event") print(ax.get_xlim()) print(ax.get_ylim()) (x_min, x_max) = ax.get_xlim() (y_min, y_max) = ax.get_ylim() x = [] y = [] z = [] for (lx, ly, lz) in zip(lon, lat, bt): if (x_min < lx < x_max) and (y_min < ly < y_max): x.append(lx) y.append(ly) z.append(lz) if sc is not None: print((len(z), min(z), max(z))) sc.set_offsets(np.c_[x, y]) fig.canvas.draw_idle() plt.pause(0.1) #sc.remove() #sc = plt.scatter(x, y, s=50, c=z, marker='o', cmap=cm, vmin=int(min(z)) , vmax=int(max(z)) + 1) #fig.canvas.draw() draw_event_cb_id = fig.canvas.mpl_connect("draw_event", on_draw) fig.set_dpi(100) fig.set_size_inches(8, 4.44, forward=True) fig.suptitle("Brightness Temperature") plt.subplots_adjust(left=0.05, right=0.90, bottom=0.05, top=0.90) plt.xlim(-180.0, 180.0) plt.ylim(-90.0, 90.0) man = plt.get_current_fig_manager() man.canvas.set_window_title(window_title) world_path = os.getenv("MWS_HOME_PAT") + "/tools/PAT/data/world/world_010m.txt" fp = open(world_path) world_lines = fp.readlines() fp.close() data_x = [] data_y = [] for line in world_lines: if len(line) == 1: if len(data_x) > 0: plt.plot(data_x, data_y, marker="", color="black", markersize=0, linewidth=1) data_x = [] data_y = [] else: x, y = line.split() data_x.append(float(x)) data_y.append(float(y)) #plt.imshow(lon, lat, bt, cmap=plt.get_cmap("rainbow")) sc = plt.scatter(lon, lat, s=50, c=bt, marker='o', cmap=cm, vmin=int(min(bt)), vmax=int(max(bt)) + 1) cax = plt.axes([0.91, 0.05, 0.02, 0.9]) plt.colorbar(cax=cax) plt.show() ------------------------------------------------------------------------------ E-MAIL DISCLAIMER The present message may contain confidential and/or legally privileged information. If you are not the intended addressee and in case of a transmission error, please notify the sender immediately and destroy this E-mail. Disclosure, reproduction or distribution of this document and its possible attachments is strictly forbidden. SPACEBEL denies all liability for incomplete, improper, inaccurate, intercepted, (partly) destroyed, lost and/or belated transmission of the current information given that unencrypted electronic transmission cannot currently be guaranteed to be secure or error free. Upon request or in conformity with formal, contractual agreements, an originally signed hard copy will be sent to you to confirm the information contained in this E-mail. SPACEBEL denies all liability where E-mail is used for private use. SPACEBEL cannot be held responsible for possible viruses that might corrupt this message and/or your computer system. -------------------------------------------------------------------------------

1 0