Mailman 3 May 2017 - Matplotlib-devel

Matplotlib Finance Maintenance
by dmcnick150＠frontier.com Aug. 5, 2017

Aug. 5, 2017

Hello! What is involved in maintaining the matplotlib.finance module? I am a daily user of the module and consider myself to be an okay Python script writer, but I wonder if maintaining the module is beyond my skill. Thanks for your consideration, David McNickle dmcnick150(a)frontier.com Sent from myMail for iOS

2 1

phone call 2017-05-22
by Thomas Caswell May 22, 2017

May 22, 2017

Just my self and Eric Firing - brief discussion about also using circleci for testing / docs building as suggested by Qulogic - Eric is -1 on adding a third CI service and would rather see effort to make the tests run faster (like only running 'slow' tests on a cron job once a day) - Tom is neutral on it - Eric reports a possible bug with Qt5 + hi-dpi widgets ( http://currents.soest.hawaii.edu/hgstage/pycurrents/file/45193e2ff8ad/plot/… <http://currents.soest.hawaii.edu/hgstage/pycurrents/file/45193e2ff8ad/plot/…> ) that sounds like bad interaction of blitting and dpi scaling. I will investigate. Tom

1 0

difference between filled and unfilled markers with mpl 2.0 style ?
by Pierre Haessig May 18, 2017

May 18, 2017

Hello, Just a small inconsistency I noticed in the doc examples "lines_bars_and_markers" which relates to the 2.0 style change: * 1.5 : http://matplotlib.org/1.5.3/examples/lines_bars_and_markers/marker_referenc… * 2.0 : http://matplotlib.org/examples/lines_bars_and_markers/marker_reference.html The difference between filled and unfilled markers (for '.' vs 'o' or '5' vs '>' and some others) has become almost invisible. At least, it not about filled vs unfilled anymore. I would recommend keeping in this "unfilled" category only the markers made of segments (like '+', x'...) and promote the others (like '.', '5', '9'...) to filled. What do you think? Now as a different but related question, I'm missing a bit marker names for "truly unfilled" squares and circles. As of now, it takes a plot(x, y, 'o', fillstyle='none') to got a hollow circle. What do you think ? (maybe I'm just a bit lazy...) (Maybe 'O' could be hollow version of 'o'? This "capital trick" could work for 's' or 'v' as well, but fail for 'd' or '<'... ) best, Pierre

1 0

Matplotlib 2.0.2
by Thomas Caswell May 16, 2017

May 16, 2017

Folks, I just tagged 2.0.2 Critical bug fixes for 2.0.1 - fixed Qt4 support - fixed LogFormatter - fixed hatched artists in legends - fixed segfault with large images I will do a wider announcement, rebuild the docs, and upload to pypi once I have collected the wheels. Tom

2 4

2017-05-08 phone call
by Thomas Caswell May 14, 2017

May 14, 2017

Folks, Sorry for being so late on getting this out. My self, Eric Firing and Ryan May. - Went over the last few issues for 2.0.2 (which has subsequently been tagged and released) - discussed future release plans Going forward, we do not plan to do a 2.0.3 (unless we find another truly egregious regression) and plan to get a 2.1 RC in time for scipy (July). This leaves the question of what to do with our milestones as we have 600+ things tagged for 2.0.3. In general, we need some way to differentiate 'This must go in for this release X.Y.Z' and 'This can / sholud be merged / backported for release X.Y.Z'. Maybe we should have a 'next minor' and 'next patch' milestones to hold the 'can/should' issues / PRs and use the explicitly numbered milestones for the 'must' category? Tom

2 1

Re: [Matplotlib-devel] Data exploration tools when plotting
by Isaac Gerg May 3, 2017

May 3, 2017

I've never heard of nexpy -- thanks for sharing! Aside from having a simple tool to explore data in a fashion similar to matlab, it would be nice to have a properties editor so that one can futz with the linestyle's etc until satisfied and then be able to update their rc file. Isaac On Wed, May 3, 2017 at 2:25 PM, Osborn, Raymond <rosborn(a)anl.gov> wrote: > It is probably overkill for what you want, but perhaps I can put a plug in > for NeXpy (http://nexpy.github.io/nexpy/), which embeds a Matplotlib > window in a PyQt GUI that includes an IPython shell. It automatically shows > the cursor positions in the Options tab. Your question makes me realize > that it would be easy to store the values of a cursor click in shell > variables - it is used programmatically in a few places, but the values are > not currently accessible from the shell. I am releasing a new version soon, > so I will see if I can include it. > > NeXpy is designed as a simple toolbox for viewing and manipulating HDF5 > files, with an underlying API that allows you to construct HDF5 data > structures in memory before saving them. However, you can do a lot in the > shell without needing to save the data as HDF5. Using your example, you > would just need to type in the shell > > >>> x=np.linspace(0, 2*np.pi, 100) > >>> NXdata(np.sin(x), x).plot() > > Ray > > On May 3, 2017, at 9:19 AM, Isaac Gerg <isaac.gerg(a)gergltd.com> wrote: > >> >>>>> On Apr 28, 2017 1:10 PM, "Isaac Gerg" <isaac.gerg(a)gergltd.com> wrote: >>>>> >>>>>> Are there any plans to put in data exploration tools into plotting? >>>>>> For example, if I do a simple plt.plot(np.sin(np.linspace(0,2*np.pi), >>>>>> 100); plt.show(), putting a tool in the toolbar whereby I can click on a >>>>>> point and get its x,y value? >>>>>> >>>>>> Thanks, >>>>>> Isaac >>>>>> >>>>>> _______________________________________________ >>>>>> Matplotlib-devel mailing list >>>>>> Matplotlib-devel(a)python.org >>>>>> https://mail.python.org/mailman/listinfo/matplotlib-devel >>>>>> >>>>>> >>>> >>>> _______________________________________________ >>>> Matplotlib-devel mailing list >>>> Matplotlib-devel(a)python.org >>>> https://mail.python.org/mailman/listinfo/matplotlib-devel >>>> >>>> >>> _______________________________________________ >>> Matplotlib-devel mailing list >>> Matplotlib-devel(a)python.org >>> https://mail.python.org/mailman/listinfo/matplotlib-devel >>> >> > _______________________________________________ > Matplotlib-devel mailing list > Matplotlib-devel(a)python.org > https://mail.python.org/mailman/listinfo/matplotlib-devel > > > -- > Ray Osborn, Senior Scientist > Materials Science Division > Argonne National Laboratory > Argonne, IL 60439, USA > Phone: +1 (630) 252-9011 <(630)%20252-9011> > Email: ROsborn(a)anl.gov > > >

1 0

Data exploration tools when plotting
by Isaac Gerg May 3, 2017

May 3, 2017

Are there any plans to put in data exploration tools into plotting? For example, if I do a simple plt.plot(np.sin(np.linspace(0,2*np.pi), 100); plt.show(), putting a tool in the toolbar whereby I can click on a point and get its x,y value? Thanks, Isaac

4 5

[REL] Matplotlib v2.0.1
by Thomas Caswell May 3, 2017

May 3, 2017

Folks, Happy to announce the only slightly delayed release of 2.0.1, the first bug fix release for 2.0.x series. Source and wheels are an pypi and conda packages are available from the conda forge channel pip install --upgrade matplotlib or conda install matplotlib -c conda-forge This release includes bug fixes, documentation updates and one major change. API - Do not clip the linewidth used for dash pattern scaling. This is an API change, but is a major improvement. - Deprecate 'vega' color names in favor of 'tab'. Bug Fixes - Hatching color follows edge color again. - Fixes a critical bug with masked images. - Improved high-dpi support for Qt5. - Log ticking no never skip minor ticks - Do not skip points surrounded by nan/inf in vector outputs. Supported versions of Python - 2.7, 3.4, 3.5, 3.6 - Patches to improve compatibility with pypy Many thanks to everyone who helped with this release! Thanks to Mathew Brett and Christoph Gohlke for building the wheels. Tom

1 0

REL: matplotlib v2.0.1
by Thomas Caswell May 2, 2017

May 2, 2017

Folks, I have tagged 2.0.1 and (attempted) to kick off the various binary building processes. Hopefully those will all be good to go tomorrow so I can upload to pypi and make a wider announcement. At Nelle's suggestion, we are skipping an rc on this release. Tom

1 0

bar plot xtick limits with dates
by Isaac Gerg May 1, 2017

May 1, 2017

Hi all, I am trying to make a plot from a pandas dataframe and i cannot for the life of me get the xlabels to stay within range of the data. Any suggestions? attached is the plot in question. The code is: plt.figure() fig = matplotlib.pyplot.gcf() fig.set_size_inches(16,9) plt.title('Number Abnormal Events Per Week By Type') plt.bar(tmp2.index, tmp2['loose'], color='C1', width=4) plt.bar(tmp2.index, tmp2['hard'], color='C2', width=4, bottom=tmp2['loose']) plt.ylabel('Count') plt.xlabel('Week') plt.legend(['l','h']) plt.xticks(rotation='vertical') ax = matplotlib.pyplot.gca() ax.xaxis.set_major_formatter(matplotlib.dates.DateFormatter('%Y-%m-%d')) ax.xaxis.set_major_locator(mdates.WeekdayLocator(byweekday=mdates.MO)) plt.savefig(r'output\abnormal.png', bbox='tight') with tmp2.index DatetimeIndex(['2017-01-02', '2017-01-09', '2017-01-16', '2017-01-23', '2017-01-30', '2017-02-06', '2017-02-13', '2017-02-20', '2017-02-27', '2017-03-06', '2017-03-13', '2017-03-20', '2017-03-27', '2017-04-03', '2017-04-10', '2017-04-17', '2017-04-24'], dtype='datetime64[ns]', name='datetime', freq='W-MON') I would like to remove the xlabels on either end of the data (12-26 and 05-01). Thanks, Isaac [image: Inline image 1]

3 5