matplotlib-users Mailing List for matplotlib

On Sun, Aug 28, 2011 at 3:03 PM, Bruce Ford <br...@cl...>wrote:

> Getting a strange result trying to divide two 3d arrays.  I am getting
> a matrix of NaNs regardless of how I divide and I can't determine why.
>
> #opened a NetCDF file using python-netcdf4
>
> var1 = nc_file.variables['var1']  ###shape = [31,181,360] with a
> values ranging from 0 - 243 and NO NaNs in the array, dtype float32
> var2 = nc_file.variables['var2']  ###shape = [31,181,360] with a
> values ranging from 0 - 4 (mostly zeros) and NO NaNs in the array,
> dtype float32
>
> np.seterr(all='ignore')  #in case problem has something do to with
> dividing by zero
>
> var1/var2  ###gives array of NaNs with shape of 31,181,360
>
> #doing the division one slice at a time doesn't help...
> for x in range(1,var1.shape[0]):
>            var[x,:,:] = var1[x,:,:]/var2[x,:,:]  ###gives array of
> NaNs with shape of 31,181,360
>
> var = np.divide(var1,var2)  ###gives array of NaNs with shape of 31,181,360
>
>
> print "<p>where max: " + np.where(var1 == np.max(var1))  #prints
> (array([28]), array([79]), array([182]))
> print var1[28,79,182]  #print 545
> print var2[28,79,182]   ##prints 6
>
> #so there are values in this location that should not result in an
> NaN.  Instead I get an entire array of NaNs
>
> What am I missing?
>
> Bruce
>
>
Just to be sure, what might appear to be all NaNs is not really.  To find
out for sure, do this:

print np.any(np.isfinite(var1/var2))

If it is True, then the array is not full of NaNs and Infs, if False, then
that would be odd.  When you print an array to the screen, it will often
only print part of the array.

Furthermore, the author of the post that you link to in your other email
misses some important facts about representation of numbers in computers.
The reason 0 / 0 (integers) did not return a NaN or 2 / 0 return an Inf is
that integers can not represent those values.  Only floating point numbers
can.  Furthermore, there is an important difference between Inf and NaN.  It
is merely out of convenience that nan_to_num() would convert the Infs into
the largest (or smallest for -Inf) floating point number.  Other functions
designed for NaN (such as nansum(), nanmax(), etc) do not treat Infs in this
manner.

Ben Root

On Mon, Aug 29, 2011 at 2:56 PM, Mark Bakker <ma...@gm...> wrote:
> Hello List,
>
> Does anybody know of any plans to include support for the new datetime64
> data type in numpy? If this is the new numpy standard for doing dates and
> times, it would be great if it would work with plot_date, for example.
>
> Just wondering (but boy, would I do a little dance when all this datetime
> stuff is fully operational and integrated),


Should be fairly straightforward to write a converter.  Just follow
the example of DateConverter in matplotlib.dates
:https://github.com/matplotlib/matplotlib/blob/master/lib/matplotlib/dates.py:

class DateConverter(units.ConversionInterface):
    """
Converter for datetime.date and datetime.datetime data,
or for date/time data represented as it would be converted
by :func:`date2num`.

The 'unit' tag for such data is None or a tzinfo instance.
"""

    @staticmethod
    def axisinfo(unit, axis):
        """
Return the :class:`~matplotlib.units.AxisInfo` for *unit*.

*unit* is a tzinfo instance or None.
The *axis* argument is required but not used.
"""
        tz = unit

        majloc = AutoDateLocator(tz=tz)
        majfmt = AutoDateFormatter(majloc, tz=tz)
        datemin = datetime.date(2000, 1, 1)
        datemax = datetime.date(2010, 1, 1)

        return units.AxisInfo( majloc=majloc, majfmt=majfmt, label='',
                               default_limits=(datemin, datemax))

    @staticmethod
    def convert(value, unit, axis):
        """
If *value* is not already a number or sequence of numbers,
convert it with :func:`date2num`.

The *unit* and *axis* arguments are not used.
"""
        if units.ConversionInterface.is_numlike(value):
            return value
        return date2num(value)

    @staticmethod
    def default_units(x, axis):
        'Return the tzinfo instance of *x* or of its first element, or None'
        try:
            x = x[0]
        except (TypeError, IndexError):
            pass

        try:
            return x.tzinfo
        except AttributeError:
            pass
        return None


units.registry[datetime.date] = DateConverter()
units.registry[datetime.datetime] = DateConverter()

Hello List,

Does anybody know of any plans to include support for the new datetime64
data type in numpy? If this is the new numpy standard for doing dates and
times, it would be great if it would work with plot_date, for example.

Just wondering (but boy, would I do a little dance when all this datetime
stuff is fully operational and integrated),

Mark

On 08/29/2011 07:40 AM, Benjamin Root wrote:
>
>
> On Mon, Aug 29, 2011 at 12:30 PM, Eric Firing <ef...@ha...
> <mailto:ef...@ha...>> wrote:
>
>     On 08/29/2011 05:44 AM, Benjamin Root wrote:
>      > On Mon, Aug 29, 2011 at 9:59 AM, James McMurray
>     <jam...@gm... <mailto:jam...@gm...>
>      > <mailto:jam...@gm... <mailto:jam...@gm...>>> wrote:
>      >
>      >     Okay, so using cm.set_over() and set_under works, but is
>     there any
>      >     way of adding the set colors to the edges of the colorbar?
>      >
>      >
>      >
>      > This example should do what you want:
>      >
>      > http://matplotlib.sourceforge.net/examples/api/colorbar_only.html
>
>     Also
>     http://matplotlib.sourceforge.net/examples/pylab_examples/contourf_demo.html.
>
>     Note that "extend=both" is the relevant kwarg.  With contourf, the kwarg
>     goes in the contourf call, and colorbar automatically gets it from the
>     ContourSet. With other types of color mappable plot, the kwarg needs to
>     be given directly to the colorbar call.
>
>     Eric
>
>
> Crap! Is that why my colorbars from my pcolors (or was it pcolormesh?)
> never worked right?  I have been making colorbars directly for a while
> now because I couldn't figure out why those colorbars didn't work right.
>
> I think I will file a feature request to fix this inconsistency.
> Probably will be a bit involved to fix it, though.

It's not entirely an inconsistency.  The logic is this: extend='both' 
(or whatever) directly affects the levels used in contourf; there is 
nothing similar in imshow or pcolor*.  Adding the extend kwarg to them 
would simply be putting it in place for later use by colorbar.

Eric

>
> Ben Root
>
>
>
> ------------------------------------------------------------------------------
> EMC VNX: the world's simplest storage, starting under $10K
> The only unified storage solution that offers unified management
> Up to 160% more powerful than alternatives and 25% more efficient.
> Guaranteed. http://p.sf.net/sfu/emc-vnx-dev2dev
>
>
>
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

On Mon, Aug 29, 2011 at 12:30 PM, Eric Firing <ef...@ha...> wrote:

> On 08/29/2011 05:44 AM, Benjamin Root wrote:
> > On Mon, Aug 29, 2011 at 9:59 AM, James McMurray <jam...@gm...
> > <mailto:jam...@gm...>> wrote:
> >
> >     Okay, so using cm.set_over() and set_under works, but is there any
> >     way of adding the set colors to the edges of the colorbar?
> >
> >
> >
> > This example should do what you want:
> >
> > http://matplotlib.sourceforge.net/examples/api/colorbar_only.html
>
> Also
>
> http://matplotlib.sourceforge.net/examples/pylab_examples/contourf_demo.html
> .
>
> Note that "extend=both" is the relevant kwarg.  With contourf, the kwarg
> goes in the contourf call, and colorbar automatically gets it from the
> ContourSet. With other types of color mappable plot, the kwarg needs to
> be given directly to the colorbar call.
>
> Eric
>
>
Crap! Is that why my colorbars from my pcolors (or was it pcolormesh?) never
worked right?  I have been making colorbars directly for a while now because
I couldn't figure out why those colorbars didn't work right.

I think I will file a feature request to fix this inconsistency.  Probably
will be a bit involved to fix it, though.

Ben Root

On 08/29/2011 05:44 AM, Benjamin Root wrote:
> On Mon, Aug 29, 2011 at 9:59 AM, James McMurray <jam...@gm...
> <mailto:jam...@gm...>> wrote:
>
>     Okay, so using cm.set_over() and set_under works, but is there any
>     way of adding the set colors to the edges of the colorbar?
>
>
>
> This example should do what you want:
>
> http://matplotlib.sourceforge.net/examples/api/colorbar_only.html

Also 
http://matplotlib.sourceforge.net/examples/pylab_examples/contourf_demo.html.

Note that "extend=both" is the relevant kwarg.  With contourf, the kwarg 
goes in the contourf call, and colorbar automatically gets it from the 
ContourSet. With other types of color mappable plot, the kwarg needs to 
be given directly to the colorbar call.

Eric

>
> I hope that helps!
> Ben Root
>
>
>
> ------------------------------------------------------------------------------
> EMC VNX: the world's simplest storage, starting under $10K
> The only unified storage solution that offers unified management
> Up to 160% more powerful than alternatives and 25% more efficient.
> Guaranteed. http://p.sf.net/sfu/emc-vnx-dev2dev
>
>
>
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Closing the loop...I found the difficulty and applied this:

var = np.nan_to_num(np.divide(var1,var2))

See this page:  http://psaffrey.wordpress.com/2010/07/30/numpy-and-nan-again/

---------------------------------------
Bruce W. Ford
Clear Science, Inc.
br...@cl...






On Sun, Aug 28, 2011 at 4:03 PM, Bruce Ford <br...@cl...> wrote:
> Getting a strange result trying to divide two 3d arrays.  I am getting
> a matrix of NaNs regardless of how I divide and I can't determine why.
>
> #opened a NetCDF file using python-netcdf4
>
> var1 = nc_file.variables['var1']  ###shape = [31,181,360] with a
> values ranging from 0 - 243 and NO NaNs in the array, dtype float32
> var2 = nc_file.variables['var2']  ###shape = [31,181,360] with a
> values ranging from 0 - 4 (mostly zeros) and NO NaNs in the array,
> dtype float32
>
> np.seterr(all='ignore')  #in case problem has something do to with
> dividing by zero
>
> var1/var2  ###gives array of NaNs with shape of 31,181,360
>
> #doing the division one slice at a time doesn't help...
> for x in range(1,var1.shape[0]):
>            var[x,:,:] = var1[x,:,:]/var2[x,:,:]  ###gives array of
> NaNs with shape of 31,181,360
>
> var = np.divide(var1,var2)  ###gives array of NaNs with shape of 31,181,360
>
>
> print "<p>where max: " + np.where(var1 == np.max(var1))  #prints
> (array([28]), array([79]), array([182]))
> print var1[28,79,182]  #print 545
> print var2[28,79,182]   ##prints 6
>
> #so there are values in this location that should not result in an
> NaN.  Instead I get an entire array of NaNs
>
> What am I missing?
>
> Bruce
>
> ---------------------------------------
> Bruce W. Ford
> Clear Science, Inc.
> br...@cl...
>

On Mon, Aug 29, 2011 at 9:59 AM, James McMurray <jam...@gm...>wrote:

> Okay, so using cm.set_over() and set_under works, but is there any way of
> adding the set colors to the edges of the colorbar?
>
>
>
This example should do what you want:

http://matplotlib.sourceforge.net/examples/api/colorbar_only.html

I hope that helps!
Ben Root

Okay, so using cm.set_over() and set_under works, but is there any way of
adding the set colors to the edges of the colorbar?

On 29 August 2011 15:47, James McMurray <jam...@gm...> wrote:

> Hi,
>
> Is it possible to color values outside the colorbar limits in a different
> color than the colorbar itself. So for example, an otherwise greyscale plot
> could have the values outside the upperlimit in blue and those below the
> lower limit in red, so that these points are not confused with the
> legitimate data points (as they can be when they are just more black and
> white points as default).
>
> Thanks,
> James McMurray
>

Can you describe what you want to do? So you now want a histogram?


surfcast23 wrote:
> 
> Sorry everyone I totally missed something very important. What I need to
> do is first bin the masses(which I don't know how to do). 
> 
> Chelonian wrote:
>> 
>> On Thu, Aug 25, 2011 at 10:01 PM, surfcast23 <sur...@gm...>
>> wrote:
>>>
>>> Hi,
>>>
>>>   there is only one column. so I want a plot of y and x. With y taking
>>> values running from 0 to n  or 7 in my example and x as the average of
>>> the
>>> values that are contained in the rows in my example it was 5.57.
>> 
>> It seems to me that, as described, you want a plot that in which all
>> the bars are the same height (or width if it is a sideways bar chart),
>> in this case, 5.57.  That makes no sense.
>> 
>> What information is this plot is intended to provide the viewer?
>> 
>> ------------------------------------------------------------------------------
>> EMC VNX: the world's simplest storage, starting under $10K
>> The only unified storage solution that offers unified management 
>> Up to 160% more powerful than alternatives and 25% more efficient. 
>> Guaranteed. http://p.sf.net/sfu/emc-vnx-dev2dev
>> _______________________________________________
>> Matplotlib-users mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>> 
>> 
> 
> 

-- 
View this message in context: http://old.nabble.com/How-do-you-Plot-data-generated-by-a-python-script--tp32328822p32357095.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

Hi,

Is it possible to color values outside the colorbar limits in a different
color than the colorbar itself. So for example, an otherwise greyscale plot
could have the values outside the upperlimit in blue and those below the
lower limit in red, so that these points are not confused with the
legitimate data points (as they can be when they are just more black and
white points as default).

Thanks,
James McMurray

Eu gostaria de adicioná-lo à minha rede profissional no LinkedIn.
-Samuel Teixeira

Samuel Teixeira Santos
Analista B na Embrapa Cerrados
Brasília e redondezas, Brasil

Confirme que você conhece Samuel Teixeira Santos:
https://www.linkedin.com/e/-b7qixr-grxfc2g9-4q/isd/4010246358/ylaRjg6e/

--
Você está recebendo convites de conexão por e-mail. Clique aqui para parar de recebê-los:
http://www.linkedin.com/e/-b7qixr-grxfc2g9-4q/_dML4GcZf3-2F5VO_wzm4ASVf0JrnPVO-ObO4eeHsZV8OPkmQwLeZN8i/goo/matplotlib-users%40lists%2Esourceforge%2Enet/20061/I1381245525_1/

(c) 2011 LinkedIn Corporation. 2029 Stierlin Ct, Mountain View, CA 94043 - EUA.

Hello
I have an animation which shows 3D diffusion
I added color bar at the side 
during the animation the colors changing but the scale at the colorbar of
the graph stay asame.
please help with that soon as you can!
Thanks!
Gal:working:
-- 
View this message in context: http://old.nabble.com/Refresh-colorbar-in-3D-graph-animation-tp32356118p32356118.html
Sent from the matplotlib - users mailing list archive at Nabble.com.