matplotlib-users Mailing List for matplotlib

Folks,

During the sprints at SciPy Mike Droettboom and I started to adapt the
Jupyter governance documents for matplotlib.

We are doing this in the open, and a pull request of the changes is at:

https://github.com/matplotlib/governance/pull/1

To be clear, this is a work in progress and is not final or in force.

We propose the mechanism for ratifying the these documents is to, after
input from the community, have the current "matplotlib Executive Committee"
(the signers of the FSA with NUMFocus) vote.

Tom

ps A very similar email was sent to the matplotlib-devel list on July 17

Dear all matplotlib users,

I am trying to make a geographical map with colors, in which the colors of
pixels will depend on 2-dimensions of (x,y), like the one attached from one
publication. I searched online and this is where I arrive:

https://gist.github.com/ChaoYue/81eb01b558f068f11ee741c56557a6a2

To put it short, I know how to pick up color by x-axis value, and use
y-axis value as an indicator of either saturation, or brightness or
transparency of the color. Or somewhere to convert this (x,y) information
back to linear space and select color from a matplotlib colormap. But is
there some way that I can use different colors for each corner of the small
square as shown in the example, and somewhere interpolate the colors in
between? like the one attached?

These are the online sources that help me, even though I am not sure I
fully understand the tricks inside them ...:

http://stackoverflow.com/questions/15207255/is-there-any-way-to-use-bivariate-colormaps-in-matplotlib

https://github.com/matplotlib/matplotlib/issues/4369

Thanks a lot in advance for any suggestions,

Cheers,
Chao

On Mon, Jul 25, 2016 at 4:34 PM, Chao YUE <cha...@gm...> wrote:

> Dear all matplotlib users,
>
> I am trying to make a geographical map with colors, in which the colors of
> pixels will depend on 2-dimensions of (x,y), like the one attached from one
> publication. I searched online and this is where I arrive:
>
> https://gist.github.com/ChaoYue/81eb01b558f068f11ee741c56557a6a2
>
> To put it short, I know how to pick up color by x-axis value, and use
> y-axis value as an indicator of either saturation, or brightness or
> transparency of the color. Or somewhere to convert this (x,y) information
> back to linear space and select color from a matplotlib colormap. But is
> there some way that I can use different colors for each corner of the small
> square as shown in the example, and somewhere interpolate the colors in
> between? like the one attached?
>
> These are the online sources that help me, even though I am not sure I
> fully understand the tricks inside them ...:
>
>
> http://stackoverflow.com/questions/15207255/is-there-any-way-to-use-bivariate-colormaps-in-matplotlib
>
> https://github.com/matplotlib/matplotlib/issues/4369
>
> Thanks a lot in advance for any suggestions,
>
> Cheers,
> Chao
>
> --
> please visit:
> http://www.globalcarbonatlas.org/
>
> ***********************************************************************************
> Chao YUE
> postdoc at LSCE
> CEA-Ormes des Merisiers, F-91191 GIF-SUR-YVETTE CEDEX
> Tel: 33 1 69 08 41 87
>
> ************************************************************************************
>



-- 
please visit:
http://www.globalcarbonatlas.org/
***********************************************************************************
Chao YUE
postdoc at LSCE
CEA-Ormes des Merisiers, F-91191 GIF-SUR-YVETTE CEDEX
Tel: 33 1 69 08 41 87
************************************************************************************

scatter does not (currently) support multiple markers in a single call.
You will have to do the group-by operation and call scatter is a loop.

If you define your cmap and norm outside of scatter, you can pass the same
instances into both calls to scatter to have mpl take care of that level of
mapping for you.

Tom

On Mon, Jul 11, 2016 at 3:32 PM Chao YUE <cha...@gm...> wrote:

> OK, after posting I realize I can pass a vector to the 's' keyword in
> plt.scatter to set the marker size, like s=[20,20,20,20,20,50,20,20,20,20].
> How about using a different marker type? Thanks!
>
> Regards,
> Chao
>
> On Mon, Jul 11, 2016 at 10:27 PM, Chao YUE <cha...@gm...> wrote:
>
>> Dear all matplotlib users,
>>
>> I have a user case that, let's say I have three series data: x,y,z.
>> I would like to make a scatter plot using (x,y) as coordinates and z as
>> the color of scatter points, using cmap keyword of plt.scatter. However, I
>> would like to highlight some specific point by using a different marker (or
>> marker size) than other points.
>>
>> ​A minimum example is like below:
>> x,y,z = np.random.randn(3,10)
>> plt.scatter(x,y,c=z,cmap=matplotlib.cm.jet)
>> plt.colorbar()​
>>
>> ​If I want to use a different marker or marker size for (x[5],y[5],z[5]),
>> how could I do that?
>> The only way I can think of is to plot again for this point using
>> plt.scatter([x[5],y[5]) but define the color by manually finding the
>> colormap ​color corresponding to z[5]. However this is quite tedious. Is
>> there a better way?
>>
>> ​Thanks a lot for any help,
>>
>> Regards,
>> Chao​
>>
>> --
>> please visit:
>> http://www.globalcarbonatlas.org/
>>
>> ***********************************************************************************
>> Chao YUE
>> postdoc at LSCE
>> CEA-Ormes des Merisiers, F-91191 GIF-SUR-YVETTE CEDEX
>> Tel: 33 1 69 08 41 87
>>
>> ************************************************************************************
>>
>
>
>
> --
> please visit:
> http://www.globalcarbonatlas.org/
>
> ***********************************************************************************
> Chao YUE
> postdoc at LSCE
> CEA-Ormes des Merisiers, F-91191 GIF-SUR-YVETTE CEDEX
> Tel: 33 1 69 08 41 87
>
> ************************************************************************************
>
> ------------------------------------------------------------------------------
> Attend Shape: An AT&T Tech Expo July 15-16. Meet us at AT&T Park in San
> Francisco, CA to explore cutting-edge tech and listen to tech luminaries
> present their vision of the future. This family event has something for
> everyone, including kids. Get more information and register today.
> http://sdm.link/attshape_______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

OK, after posting I realize I can pass a vector to the 's' keyword in
plt.scatter to set the marker size, like s=[20,20,20,20,20,50,20,20,20,20].
How about using a different marker type? Thanks!

Regards,
Chao

On Mon, Jul 11, 2016 at 10:27 PM, Chao YUE <cha...@gm...> wrote:

> Dear all matplotlib users,
>
> I have a user case that, let's say I have three series data: x,y,z.
> I would like to make a scatter plot using (x,y) as coordinates and z as
> the color of scatter points, using cmap keyword of plt.scatter. However, I
> would like to highlight some specific point by using a different marker (or
> marker size) than other points.
>
> ​A minimum example is like below:
> x,y,z = np.random.randn(3,10)
> plt.scatter(x,y,c=z,cmap=matplotlib.cm.jet)
> plt.colorbar()​
>
> ​If I want to use a different marker or marker size for (x[5],y[5],z[5]),
> how could I do that?
> The only way I can think of is to plot again for this point using
> plt.scatter([x[5],y[5]) but define the color by manually finding the
> colormap ​color corresponding to z[5]. However this is quite tedious. Is
> there a better way?
>
> ​Thanks a lot for any help,
>
> Regards,
> Chao​
>
> --
> please visit:
> http://www.globalcarbonatlas.org/
>
> ***********************************************************************************
> Chao YUE
> postdoc at LSCE
> CEA-Ormes des Merisiers, F-91191 GIF-SUR-YVETTE CEDEX
> Tel: 33 1 69 08 41 87
>
> ************************************************************************************
>



-- 
please visit:
http://www.globalcarbonatlas.org/
***********************************************************************************
Chao YUE
postdoc at LSCE
CEA-Ormes des Merisiers, F-91191 GIF-SUR-YVETTE CEDEX
Tel: 33 1 69 08 41 87
************************************************************************************

Dear all matplotlib users,

I have a user case that, let's say I have three series data: x,y,z.
I would like to make a scatter plot using (x,y) as coordinates and z as the
color of scatter points, using cmap keyword of plt.scatter. However, I
would like to highlight some specific point by using a different marker (or
marker size) than other points.

​A minimum example is like below:
x,y,z = np.random.randn(3,10)
plt.scatter(x,y,c=z,cmap=matplotlib.cm.jet)
plt.colorbar()​

​If I want to use a different marker or marker size for (x[5],y[5],z[5]),
how could I do that?
The only way I can think of is to plot again for this point using
plt.scatter([x[5],y[5]) but define the color by manually finding the
colormap ​color corresponding to z[5]. However this is quite tedious. Is
there a better way?

​Thanks a lot for any help,

Regards,
Chao​

-- 
please visit:
http://www.globalcarbonatlas.org/
***********************************************************************************
Chao YUE
postdoc at LSCE
CEA-Ormes des Merisiers, F-91191 GIF-SUR-YVETTE CEDEX
Tel: 33 1 69 08 41 87
************************************************************************************

I originally posted to Stack Overflow at
http://stackoverflow.com/q/38274681/2988730.

I am trying to follow the answer at
http://stackoverflow.com/a/38008501/2988730 to an earlier question of
mine to create colored and styled legend-like entries. I have the
following code:

    import matplotlib as mpl
    mpl.use('ps')
    from matplotlib import pyplot as plt

    mpl.rc('text', usetex=True)
    mpl.rc('text.latex', preamble='\\usepackage{color}\n\\usepackage{dashrule}')

    plt.ion()
    ax = plt.plot((0, 1), (1, 2))[0].axes
    ax.set_ylabel(r'Y $\;$ \textcolor[rgb]{1.0, 0.0,
0.0}{\hdashrule[0.5ex]{3cm}{1pt}{1pt 0pt}}')
    ax.set_xlabel(r'N $\;$ \textcolor[rgb]{0.0, 1.0,
0.0}{\rule[0.5ex]{3cm}{1pt}}')
    plt.savefig('test.ps')

The result is as expected. The labels contain black text with a red
line on the Y label and a green line on the X label:
http://i.stack.imgur.com/JCiLI.png.

However, when I try the exact same set of commands without the
`mpl.use('ps')` line (using `'qt4agg'` backend on my system), the
figure neither saves corectly nor shows up correctly on screen:

    import matplotlib as mpl
    from matplotlib import pyplot as plt

    mpl.rc('text', usetex=True)
    mpl.rc('text.latex', preamble='\\usepackage{color}\n\\usepackage{dashrule}')

    plt.ion()
    ax = plt.plot((0, 1), (1, 2))[0].axes
    ax.set_ylabel(r'Y $\;$ \textcolor[rgb]{1.0, 0.0,
0.0}{\hdashrule[0.5ex]{3cm}{1pt}{1pt 0pt}}')
    ax.set_xlabel(r'N $\;$ \textcolor[rgb]{0.0, 1.0,
0.0}{\rule[0.5ex]{3cm}{1pt}}')

    plt.savefig('test.png')
    plt.show()

The result of `plt.savefig` (http://i.stack.imgur.com/h2LXn.png) and
`plt.show` (http://i.stack.imgur.com/0Ow7c.png) are basically the same
in this case. The lines after the text show up black.

How do I get the colors to show up in the labels with the default
interactive backend?

Regards,

    -Joe