matplotlib-users Mailing List for matplotlib

> Thanks for clarifying. I understand what you're saying now. I think
> what we want to do is store the unmultiplied alpha as a "canonical"
> version of the image, and premultiply a copy (or use some C++ iterator
> magic to avoid the copy) right before sending it off to Agg. Then the
> alpha can be fully "tweakable" at runtime.
Just for my understanding of the design approach of matplotlib in
general, is it implicitly assumed that once pixels have been passed to
agg, they are destined for the renderer? No other possible use? And
the life span is short? I'm just not very familiar with where _image
is used throughout the rest of the code, if at all.
As far as the iterator magic, the kicker is that it would have to take
place inside of agg; it cannot be on the mpl side;
So if I'm understanding correctly,
 1) create a copy of the image within resize()
 2) premultiply the alpha on the copy
 3) let agg do its work, returning its result in a new pixmap
(resized) that is premultiplied.
 4) toss the extra copy made in (1)
 5) in each individual backend's renderer, right before the pixels
are rendered, unmultiply the alpha if necessary (as is the case with
wxagg).
I also remember you making mention of saving pngs above, I suppose
there would have to be an unmultiply there as well.
i've done some grepping through the code, and don't see that these
filter routines are used anywhere except specifically in this resize()
function. I can fully understand the desire to comply with agg's
spec, but I'm still not convinced that a small patch to agg isn't
appropriate here. It's a little odiferous, but won't be hard to
maintain given that agg is a very slowly changing library (last
release 2006?). Or maybe treat patching agg as short term solution.
Looking through the filter algorithms too, it looks like exactly the
same ops are done to each channel (RGBA) separately. So (it appears
as if, to my untrained eye) that the algorithms are applicable
regardless of whether or not the alphas are premultiplied. Further,
removing the clip will have very minimal impact on "proper" usage of
agg (i.e. sending in premultiplied alpha).
> I'll try to tackle this problem, as well as the problem that set_alpha
> simply doesn't work, at the same time when I get a chance (or patches
> are always welcome, of course :).
I have a patch, but you don't like it :D
But seriously, I'll take a more careful look at what it would take to
implement items 1-5 above; it might not be so bad, but I'm
uncomfortable dealing with the "underbelly" of matplotlib, as I'm not
familiar with either mpl's design nor the general subject of graphics.

I have a simple fix for this on this branch:
https://github.com/mdboom/matplotlib/tree/slow_update
It's sort of the simplest thing that could work. It caches the last 
results of "_update_ticks" and only updates them if the view limits or 
axis position have changed. It also invalidates this cache upon 
changing the set of ticks etc.
It makes things considerably faster -- it still has to draw all of the 
elements of the plot every time. Fixing that would require a lot more 
low-level replumbing (matplotlib doesn't currently do region-based 
updating etc.).
Mike
On 10/20/2011 11:17 PM, John Gu wrote:
> Thanks. Is there a place where these sorts of issues can be submitted 
> for review / fixes? I'm totally willing to take a look at possible 
> solutions if someone points me in the right direction. Thanks.
>
> On Thu, Oct 20, 2011 at 10:52 PM, Eric Firing <ef...@ha... 
> <mailto:ef...@ha...>> wrote:
>
> On 10/20/2011 03:47 PM, John Gu wrote:
> > Hello,
> >
> > I'm using version 1.0.1 of matplotlib on a linux machine. uname -a
> > returns the following: Linux jgulinux 2.6.35.14-95.fc14.x86_64
> #1 SMP
> > Tue Aug 16 21:01:58 UTC 2011 x86_64 x86_64 x86_64 GNU/Linux.
> >
> > [jgu@jgulinux ~/.matplotlib]$ cat matplotlibrc
> > backend : TkAgg
> > interactive : True
> >
> > Program that reproduces the problem (run in ipython):
> >
> > x = arange(1000)
> > y = x
> >
> > f = figure(1)
> > for ia in xrange(5):
> > for ib in xrange(5):
> > ax = subplot(5,5,ia*5+ib+1, navigate=True)
> > ax.plot(x,y)
> >
> > If we now tried to interactively navigate through any single
> subplot of
> > the figure, it is extremely slow. Not sure if this is a
> configuration
> > error on my side?
>
> No, and it is not backend-dependent, either. And it does not even
> require that all of the plots have data--it is the same if only
> the last
> subplot has data. A little experimentation indicates that all the
> ticks
> and tick labels on the other subplots are slowing down the interaction
> with the one subplot being navigated.
>
> ax.tick_params(labelbottom=False, labelleft=False)
> ax.tick_params(left=False, right=False, top=False,
> bottom=False)
>
> Insert the above to turn off ticks and tick labels, and it speeds up
> nicely. That's not a solution, I realize.
>
> I knew that slowness in the tick and tick label generation was a major
> bottleneck in subplot creation, but I did not realize that it had such
> an effect on interactive manipulation of a single subplot.
>
> Eric
>
> >
> > John
>
> ------------------------------------------------------------------------------
> The demand for IT networking professionals continues to grow, and the
> demand for specialized networking skills is growing even more rapidly.
> Take a complimentary Learning@Cisco Self-Assessment and learn
> about Cisco certifications, training, and career opportunities.
> http://p.sf.net/sfu/cisco-dev2dev
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> <mailto:Mat...@li...>
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>
>
> ------------------------------------------------------------------------------
> The demand for IT networking professionals continues to grow, and the
> demand for specialized networking skills is growing even more rapidly.
> Take a complimentary Learning@Cisco Self-Assessment and learn
> about Cisco certifications, training, and career opportunities.
> http://p.sf.net/sfu/cisco-dev2dev
>
>
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

On Friday, October 21, 2011, Michael Droettboom <md...@st...> wrote:
> On 10/21/2011 09:49 AM, Daniel Hyams wrote:
>> All sounds reasonable Mike; I do agree that patching the agg source
>> code is not that desirable; I was operating under the (incorrect)
>> assumption that most, if not all, backends used straight alpha.
>>
>> I'll certainly test the patch tonight, but I can only test it under
>> wxAgg reasonably, which is one of the backends that expects straight
>> alpha as far as I know.
>>
>> The "loss in picture detail" comment was just discomfort with the fact
>> that once the alphas are premultiplied in, there is not an exact
>> reverse transformation to get your original color back. In googling
>> around to better explain here, I found this, which is a much more in
>> depth and better explanation than what I would have come up with:
>>
>> http://www.quasimondo.com/archives/000665.php
>>
>> The crux is here:
>>
>>> An example: when you set the alpha value of a pixel to 16 all color
values will be multiplied with
>>> a factor of 16/256 = 0.0625. So a gray pixel of 128 will become 128 *
0.0625 = 8, a darker pixel
>>> of 64 will become 64 * 0.0625 = 4. But a slightly lighter pixel of maybe
67 will become 67 *
>>> 0.0625 = 4.1875 - yet there are no decimals in integer pixels which
means it will also become
>>> 4. The effect that you will get posterization - setting your alpha
channel to 8 means that you
>>> also reduce your color channels to 8 levels, this means instead =
256*256*256 different colors
>>> you will end up with a maximum of 8*8*8 = 512 different colors.
>> One might argue that the loss of color information is not that
>> crucial, because for very low alpha (where the problem is most
>> pronounced), the image is almost invisible anyway... so it won't
>> matter. That's true, but what if I want to (for whatever reason) take
>> the image's pixels again, before draw, and boost the alpha up again?
>> What I'll get is a posterized mess. So, I'm still of the opinion that
>> patching agg in this situation might be the best solution to this.
>> This way, straight alphas are used throughout, and for backends that
>> require premultiplied alpha, the alpha can be premultiplied in at the
>> latest possible moment.
>
> Thanks for clarifying. I understand what you're saying now. I think
> what we want to do is store the unmultiplied alpha as a "canonical"
> version of the image, and premultiply a copy (or use some C++ iterator
> magic to avoid the copy) right before sending it off to Agg. Then the
> alpha can be fully "tweakable" at runtime.
>
> I'll try to tackle this problem, as well as the problem that set_alpha
> simply doesn't work, at the same time when I get a chance (or patches
> are always welcome, of course :).
>
> Mike
Mike,
This idea is sort of along the same vein of the idea that I have been having
for unifying the colors framework and allowing for transforms to be
performed upon access of the colors. If you think about it, the issue of
different backends wanting different information can be dealt with by having
a transform that returns pre multiplied alphas and another that returns
straight alphas. The backends can then solely the desired transform that it
expects.
I guess the question is whether the transformation step should be
generalized at the python level, or specialized down at the python/c++
boundary.
I really wish I had more time this semester to try and come up with a proof
of concept.
Ben Root
P.S. - I had also noticed alpha-blending issues before (and asked about it),
but I ended up convincing myself (apparently wrongly) that it was correct.

On 10/21/2011 09:49 AM, Daniel Hyams wrote:
> All sounds reasonable Mike; I do agree that patching the agg source
> code is not that desirable; I was operating under the (incorrect)
> assumption that most, if not all, backends used straight alpha.
>
> I'll certainly test the patch tonight, but I can only test it under
> wxAgg reasonably, which is one of the backends that expects straight
> alpha as far as I know.
>
> The "loss in picture detail" comment was just discomfort with the fact
> that once the alphas are premultiplied in, there is not an exact
> reverse transformation to get your original color back. In googling
> around to better explain here, I found this, which is a much more in
> depth and better explanation than what I would have come up with:
>
> http://www.quasimondo.com/archives/000665.php
>
> The crux is here:
>
>> An example: when you set the alpha value of a pixel to 16 all color values will be multiplied with
>> a factor of 16/256 = 0.0625. So a gray pixel of 128 will become 128 * 0.0625 = 8, a darker pixel
>> of 64 will become 64 * 0.0625 = 4. But a slightly lighter pixel of maybe 67 will become 67 *
>> 0.0625 = 4.1875 - yet there are no decimals in integer pixels which means it will also become
>> 4. The effect that you will get posterization - setting your alpha channel to 8 means that you
>> also reduce your color channels to 8 levels, this means instead = 256*256*256 different colors
>> you will end up with a maximum of 8*8*8 = 512 different colors.
> One might argue that the loss of color information is not that
> crucial, because for very low alpha (where the problem is most
> pronounced), the image is almost invisible anyway... so it won't
> matter. That's true, but what if I want to (for whatever reason) take
> the image's pixels again, before draw, and boost the alpha up again?
> What I'll get is a posterized mess. So, I'm still of the opinion that
> patching agg in this situation might be the best solution to this.
> This way, straight alphas are used throughout, and for backends that
> require premultiplied alpha, the alpha can be premultiplied in at the
> latest possible moment.
Thanks for clarifying. I understand what you're saying now. I think 
what we want to do is store the unmultiplied alpha as a "canonical" 
version of the image, and premultiply a copy (or use some C++ iterator 
magic to avoid the copy) right before sending it off to Agg. Then the 
alpha can be fully "tweakable" at runtime.
I'll try to tackle this problem, as well as the problem that set_alpha 
simply doesn't work, at the same time when I get a chance (or patches 
are always welcome, of course :).
Mike
> Thanks for all of the help with this Mike,
>
> Daniel
>
> On Fri, Oct 21, 2011 at 9:31 AM, Michael Droettboom<md...@st...> wrote:
>> Thanks for looking into this deeper.
>>
>> Agg requires image buffers to be premultiplied, as described in the third
>> bullet point here. (It's not exactly clear, to say the least, but that's
>> what I take it to mean, and also from reading the code).
>>
>> http://www.antigrain.com/news/release_notes/v22.agdoc.html
>>
>> The bug is that in _image.cpp the input buffers are not declared as
>> premultiplied in _image.cpp. Arguably it is a bug that agg doesn't reject
>> filtering unmultiplied images, since the note states that the assumption is
>> that they are premultiplied by the time they get to the filters.
>>
>> I have attached a patch that fixes this. Would you mind testing it and let
>> me know how it works for you?
>>
>> On 10/20/2011 10:29 PM, Daniel Hyams wrote:
>>> I've looked all over the place through both the Python and C code, and
>>> I don't see any premultiplication of alphas at any stage before the
>>> pixels are passed off to agg, and neither can I find any place where
>>> the alphas are "unmultiplied" on the way back from agg to the backend
>>> for rendering.
>> matplotlib's support for alpha blending of images is basically by accident,
>> so it's not surprising the details aren't right.
>>
>> I think it is a bug that after reading images in we don't premultiply them
>> before sending them to Agg. That bug has existed for a long time in
>> matplotlib because no one is really using alpha images a great deal.
>> (Masked images, yes, but that implies alpha is strictly 0 or 255 and thus
>> these issues don't come into play.)
>>
>> Unmultiplying is not always necessary. Many of the GUI backends also expect
>> premultiplied alpha (Qt for example). However, there is certainly a bug in
>> writing PNG files (where the file format specifies unmultiplied).
>>
>>> It's very possible that I missed it, but I would have to miss it in
>>> two places (premultiply and the unmultiply). It looks to me like the
>>> output from agg ends up getting passed on directly to the renderer,
>>> which as far as I know, just uses straight alpha. The WxAgg renderer,
>>> for example, just creates a wx.Bitmap out of the pixels and blits it.
>>> Which means that any image going through agg's filters will not be
>>> correct if it has any pixels with alpha != 0 or != 255.
>>>
>>> [Using PIL images because they are simple to talk about...but the PIL
>>> image could alternatively be an image.py image]
>>>
>>> As far as I can tell, the image pixels current go through a pipeline
>>> like the following:
>>>
>>> [1] PIL Image -> _image image -> agg operations -> modified and/or
>>> resized _image image -> renderer
>>>
>>> If agg expects premultiplied alpha, the procedure should look something
>>> like:
>>>
>>> [2] PIL Image -> _image image -> premultiply alphas ->agg options ->
>>> unmultiply alphas -> modified and/or resized _image image -> renderer
>>>
>>> I personally don't like pipeline [2] because picture detail is lost in
>>> the "unmultiply alphas" stage. Better to use straight alpha all the
>>> way through.
>> I think what needed is:
>>
>> [3] PIL Image (or _png.cpp) -> premultiply alphas -> _image image -> agg
>> options ->
>> -> modified and/or resized _image image -> renderer -> (unmultiply
>> alphas)? -> GUI library
>>
>> That is -- all image data should be kept premultiplied internally in all
>> buffers for efficiency and because this is what Agg is designed for.
>>
>> Can you explain what you mean by "picture detail is lost in the unmultiply
>> alphas stage". There is the usual problem that by premultiplying you lose
>> any color data where alpha = 0 (and you lose resolution everywhere else, but
>> not resolution you can actually see after compositing).
>>> So long as matplotlib is using only a subset of agg algorithms that
>>> work no matter whether the alphas are premultiplied or not, I would
>>> think that the most reasonable route was the one that I took; to
>>> always pass straight alphas (sticking with pipeline [1]), and modify
>>> the agg source slightly to fit matplotlib's approach (i.e., remove the
>>> clipping there).
>> I'd be really wary of modifying agg like this. Those things become hard to
>> maintain. I think this instead a bug in matplotlib and should be fixed
>> there.
>>
>> I've put an issue in the issue tracker here:
>>
>> https://github.com/matplotlib/matplotlib/issues/545
>>
>> Cheers,
>> Mike
>>
>>> I hope that I'm not way off base (I have a sneaking feeling that I am
>>> :O ), and hope this helps. I've verified on both Linux and Windows
>>> that removing the alpha-clip lines from agg_span_image_filter_rgba.h,
>>> rebuilding matplotlib, and replacing _image.so/_image.pyd and
>>> _backend_agg.so/_backend_agg.pyd does the trick (along with passing
>>> straight alphas). So far, I've seen no ill effects on any of my
>>> plots, but I'm also not in a position to run the pixel-by-pixel
>>> comparison matplotlib tests.
>>>
>>>
>>> On Wed, Oct 19, 2011 at 7:26 PM, Daniel Hyams<dh...@gm...> wrote:
>>>> There has to be something else in play here. I'll try to keep this
>>>> short, but the summary is this: I can get the transparency to look
>>>> right, but only if 1) I put "straight" alpha in image, not
>>>> premultiplied, and 2) I hack agg to remove specificially every
>>>> instance of the lines of code that you refer to above.
>>>>
>>>> Why this is, I don't know. Hopefully I'm still misusing something.
>>>> However, it behaves as if the clipping of alpha in the agg library is
>>>> corrupting the alpha channel. I also submit that I could have broken
>>>> some other transparency capabilities of matplotlib, because I don't
>>>> know what other routines use what I hacked....I did check a few
>>>> transparent polygons and such though, and everything seemed to be
>>>> fine.
>>>>
>>>> I know that the agg library has been around for quite a long time, so
>>>> that also means that such a basic bug is unlikely.
>>>>
>>>> I've reattached the (slightly modified) script that reproduces the
>>>> problem, along with a sample image that it uses. The only change to
>>>> the script is right at the top, where a different image is read, a
>>>> quick statement is placed to add an alpha channel if there is not
>>>> already one, and I'm attempting to use premultiplied alphas. I've
>>>> also attached a screenshot of the output. Notice that in this case,
>>>> both "transparent" images look wrong.
>>>>
>>>> Now, if I 1) hack agg to remove the alpha clipping, and 2) modify the
>>>> one line in the attached python script so that I use straight alpha,
>>>> everything looks right. Specifically, I removed every instance of the
>>>> code below from xxxx, rebuilt all of the matplotlib .so's, and
>>>> specifically replaced _image.so and _backend_agg.so in my matplotlib
>>>> distribution.
>>>>
>>>> if(fg[order_type::A]> base_mask) fg[order_type::A]
>>>> = base_mask;
>>>> if(fg[order_type::R]> fg[order_type::A])
>>>> fg[order_type::R] = fg[order_type::A];
>>>> if(fg[order_type::G]> fg[order_type::A])
>>>> fg[order_type::G] = fg[order_type::A];
>>>> if(fg[order_type::B]> fg[order_type::A])
>>>> fg[order_type::B] = fg[order_type::A];
>>>>
>>>>
>>>>
>>>>
>>>> On Wed, Oct 19, 2011 at 2:34 PM, Daniel Hyams<dh...@gm...> wrote:
>>>>> Ah, thanks so much Michael! That explanation helps a great deal; I
>>>>> was always considering things in "straight alpha" format, not even
>>>>> knowing that there was alternative.
>>>>>
>>>>> I'll play with this tonight; I don't see any problem getting the thing
>>>>> working, though, now that I know what agg expects to see...
>>>>>
>>>>> And yes, alpha support in the image class would be very helpful ;)
>>>>>
>>>>> On Wed, Oct 19, 2011 at 2:16 PM, Michael Droettboom<md...@st...>
>>>>> wrote:
>>>>>> You are right that Agg is doing the resizing here. Agg expects
>>>>>> premultiplied alpha. See [1] for information about what that means.
>>>>>>
>>>>>> [1] http://en.wikipedia.org/wiki/Alpha_compositing
>>>>>>
>>>>>> After Agg interpolates the pixel values, to prevent oversaturation it
>>>>>> truncates all values to be less than alpha (which makes sense if
>>>>>> everything
>>>>>> is assumed to be premultiplied alpha). Arguably, the bug here is that
>>>>>> nearest neighbor (which doesn't have to do any blending) doesn't
>>>>>> perform the
>>>>>> truncation step -- then both would look "wrong".
>>>>>>
>>>>>> It happens in this code snippet in span_image_filter_rgba: (base_mask
>>>>>> is
>>>>>> 255)
>>>>>>
>>>>>> if(fg[order_type::A]> base_mask)
>>>>>> fg[order_type::A]
>>>>>> = base_mask;
>>>>>> if(fg[order_type::R]> fg[order_type::A])
>>>>>> fg[order_type::R]
>>>>>> = fg[order_type::A];
>>>>>> if(fg[order_type::G]> fg[order_type::A])
>>>>>> fg[order_type::G]
>>>>>> = fg[order_type::A];
>>>>>> if(fg[order_type::B]> fg[order_type::A])
>>>>>> fg[order_type::B]
>>>>>> = fg[order_type::A];
>>>>>>
>>>>>> So, the solution to make a partially transparent image is to not do:
>>>>>>
>>>>>> pix[:,:,3] = 127
>>>>>>
>>>>>> but instead, do
>>>>>>
>>>>>> pix *= 0.5
>>>>>>
>>>>>> Of course, the real fix here is to support alpha blending properly in
>>>>>> the
>>>>>> image class, then the user wouldn't have to deal with such details. A
>>>>>> bug
>>>>>> should probably be filed in the matplotlib issue tracker for this.
>>>>>>
>>>>>> Mike
>>>>>>
>>>>>> On 10/19/2011 12:23 PM, Daniel Hyams wrote:
>>>>>>
>>>>>> [Sorry, I keep getting tripped up with HTML mail....resent in ascii,
>>>>>> and resaved one of the attachment png's to make it smaller.]
>>>>>>
>>>>>>
>>>>>> Example script attached (PIL required). Basically, if I impose a
>>>>>> specific value into an image's alpha channel and use any interpolation
>>>>>> scheme other than 'nearest', there appears gray all where the figure
>>>>>> didn't have any color to begin with. I've also attached a screenshot
>>>>>> of the output of the script on my machine.
>>>>>>
>>>>>> Hopefully I'm doing something wrongly?
>>>>>>
>>>>>> I chased the problem and managed to hack in a solution that fixes the
>>>>>> problem, but it's extremely inefficient...basically, in matplotlib's
>>>>>> image.py, routine BboxImage.make_image, you can create two images
>>>>>> there....one with no alpha channel (call it imRGB) and one with (call
>>>>>> it imRGBA). Go through all of the routine, doing exactly the same
>>>>>> things to both of the images *except* for the interpolation, which is
>>>>>> set to 'nearest' for imRGBA. Then, rip the colors out of imRGB, the
>>>>>> alpha channel off of imRGBA, and put them together....go through all
>>>>>> of the routine again with this composited image, and it works. I
>>>>>> know...I told you it was bad ;)
>>>>>>
>>>>>> The problem seems to be in the "resize" call in that routine...resize,
>>>>>> which calls into C code, does not appear to handle things correctly
>>>>>> when the alpha is anything other than 255's across the board. It
>>>>>> might be a problem in the agg routines, but hopefully it is just maybe
>>>>>> a misuse of the agg routines.
>>>>>>
>>>>>> The behavior seems to be backend independent as far as I could test (I
>>>>>> tried with wxagg and tk backends). I am using mpl 1.0.0 on Windows if
>>>>>> it matters.
>>>>>>
>>>>>>
>>>>>> --
>>>>>> Daniel Hyams
>>>>>> dh...@gm...
>>>>>>
>>>>>>
>>>>>> ------------------------------------------------------------------------------
>>>>>> All the data continuously generated in your IT infrastructure contains
>>>>>> a
>>>>>> definitive record of customers, application performance, security
>>>>>> threats, fraudulent activity and more. Splunk takes this data and makes
>>>>>> sense of it. Business sense. IT sense. Common sense.
>>>>>> http://p.sf.net/sfu/splunk-d2d-oct
>>>>>>
>>>>>> _______________________________________________
>>>>>> Matplotlib-users mailing list
>>>>>> Mat...@li...
>>>>>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>>>>>
>>>>>>
>>>>>>
>>>>>> ------------------------------------------------------------------------------
>>>>>> The demand for IT networking professionals continues to grow, and the
>>>>>> demand for specialized networking skills is growing even more rapidly.
>>>>>> Take a complimentary Learning@Ciosco Self-Assessment and learn
>>>>>> about Cisco certifications, training, and career opportunities.
>>>>>> http://p.sf.net/sfu/cisco-dev2dev
>>>>>> _______________________________________________
>>>>>> Matplotlib-users mailing list
>>>>>> Mat...@li...
>>>>>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>>>>>
>>>>>>
>>>>>
>>>>> --
>>>>> Daniel Hyams
>>>>> dh...@gm...
>>>>
>>>> --
>>>> Daniel Hyams
>>>> dh...@gm...
>>>>
>>>
>>
>> ------------------------------------------------------------------------------
>> The demand for IT networking professionals continues to grow, and the
>> demand for specialized networking skills is growing even more rapidly.
>> Take a complimentary Learning@Cisco Self-Assessment and learn
>> about Cisco certifications, training, and career opportunities.
>> http://p.sf.net/sfu/cisco-dev2dev
>> _______________________________________________
>> Matplotlib-users mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>
>>
>
>

Thanks.
The annotate function works. Does anybody knows a way to add some text
above or belove an arrow? Now I do it using text command and
coordinates for text. May be it possible to do directly with annotate
command?
Thanks.
Petro.

All sounds reasonable Mike; I do agree that patching the agg source
code is not that desirable; I was operating under the (incorrect)
assumption that most, if not all, backends used straight alpha.
I'll certainly test the patch tonight, but I can only test it under
wxAgg reasonably, which is one of the backends that expects straight
alpha as far as I know.
The "loss in picture detail" comment was just discomfort with the fact
that once the alphas are premultiplied in, there is not an exact
reverse transformation to get your original color back. In googling
around to better explain here, I found this, which is a much more in
depth and better explanation than what I would have come up with:
http://www.quasimondo.com/archives/000665.php
The crux is here:
>An example: when you set the alpha value of a pixel to 16 all color values will be multiplied with
>a factor of 16/256 = 0.0625. So a gray pixel of 128 will become 128 * 0.0625 = 8, a darker pixel
>of 64 will become 64 * 0.0625 = 4. But a slightly lighter pixel of maybe 67 will become 67 *
> 0.0625 = 4.1875 - yet there are no decimals in integer pixels which means it will also become
> 4. The effect that you will get posterization - setting your alpha channel to 8 means that you
> also reduce your color channels to 8 levels, this means instead = 256*256*256 different colors
> you will end up with a maximum of 8*8*8 = 512 different colors.
One might argue that the loss of color information is not that
crucial, because for very low alpha (where the problem is most
pronounced), the image is almost invisible anyway... so it won't
matter. That's true, but what if I want to (for whatever reason) take
the image's pixels again, before draw, and boost the alpha up again?
What I'll get is a posterized mess. So, I'm still of the opinion that
patching agg in this situation might be the best solution to this.
This way, straight alphas are used throughout, and for backends that
require premultiplied alpha, the alpha can be premultiplied in at the
latest possible moment.
Thanks for all of the help with this Mike,
Daniel
On Fri, Oct 21, 2011 at 9:31 AM, Michael Droettboom <md...@st...> wrote:
> Thanks for looking into this deeper.
>
> Agg requires image buffers to be premultiplied, as described in the third
> bullet point here. (It's not exactly clear, to say the least, but that's
> what I take it to mean, and also from reading the code).
>
> http://www.antigrain.com/news/release_notes/v22.agdoc.html
>
> The bug is that in _image.cpp the input buffers are not declared as
> premultiplied in _image.cpp. Arguably it is a bug that agg doesn't reject
> filtering unmultiplied images, since the note states that the assumption is
> that they are premultiplied by the time they get to the filters.
>
> I have attached a patch that fixes this. Would you mind testing it and let
> me know how it works for you?
>
> On 10/20/2011 10:29 PM, Daniel Hyams wrote:
>>
>> I've looked all over the place through both the Python and C code, and
>> I don't see any premultiplication of alphas at any stage before the
>> pixels are passed off to agg, and neither can I find any place where
>> the alphas are "unmultiplied" on the way back from agg to the backend
>> for rendering.
>
> matplotlib's support for alpha blending of images is basically by accident,
> so it's not surprising the details aren't right.
>
> I think it is a bug that after reading images in we don't premultiply them
> before sending them to Agg. That bug has existed for a long time in
> matplotlib because no one is really using alpha images a great deal.
> (Masked images, yes, but that implies alpha is strictly 0 or 255 and thus
> these issues don't come into play.)
>
> Unmultiplying is not always necessary. Many of the GUI backends also expect
> premultiplied alpha (Qt for example). However, there is certainly a bug in
> writing PNG files (where the file format specifies unmultiplied).
>
>> It's very possible that I missed it, but I would have to miss it in
>> two places (premultiply and the unmultiply). It looks to me like the
>> output from agg ends up getting passed on directly to the renderer,
>> which as far as I know, just uses straight alpha. The WxAgg renderer,
>> for example, just creates a wx.Bitmap out of the pixels and blits it.
>> Which means that any image going through agg's filters will not be
>> correct if it has any pixels with alpha != 0 or != 255.
>>
>> [Using PIL images because they are simple to talk about...but the PIL
>> image could alternatively be an image.py image]
>>
>> As far as I can tell, the image pixels current go through a pipeline
>> like the following:
>>
>> [1] PIL Image -> _image image -> agg operations -> modified and/or
>> resized _image image -> renderer
>>
>> If agg expects premultiplied alpha, the procedure should look something
>> like:
>>
>> [2] PIL Image -> _image image -> premultiply alphas ->agg options ->
>> unmultiply alphas -> modified and/or resized _image image -> renderer
>>
>> I personally don't like pipeline [2] because picture detail is lost in
>> the "unmultiply alphas" stage. Better to use straight alpha all the
>> way through.
>
> I think what needed is:
>
> [3] PIL Image (or _png.cpp) -> premultiply alphas -> _image image -> agg
> options ->
> -> modified and/or resized _image image -> renderer -> (unmultiply
> alphas)? -> GUI library
>
> That is -- all image data should be kept premultiplied internally in all
> buffers for efficiency and because this is what Agg is designed for.
>
> Can you explain what you mean by "picture detail is lost in the unmultiply
> alphas stage". There is the usual problem that by premultiplying you lose
> any color data where alpha = 0 (and you lose resolution everywhere else, but
> not resolution you can actually see after compositing).
>>
>> So long as matplotlib is using only a subset of agg algorithms that
>> work no matter whether the alphas are premultiplied or not, I would
>> think that the most reasonable route was the one that I took; to
>> always pass straight alphas (sticking with pipeline [1]), and modify
>> the agg source slightly to fit matplotlib's approach (i.e., remove the
>> clipping there).
>
> I'd be really wary of modifying agg like this. Those things become hard to
> maintain. I think this instead a bug in matplotlib and should be fixed
> there.
>
> I've put an issue in the issue tracker here:
>
> https://github.com/matplotlib/matplotlib/issues/545
>
> Cheers,
> Mike
>
>> I hope that I'm not way off base (I have a sneaking feeling that I am
>> :O ), and hope this helps. I've verified on both Linux and Windows
>> that removing the alpha-clip lines from agg_span_image_filter_rgba.h,
>> rebuilding matplotlib, and replacing _image.so/_image.pyd and
>> _backend_agg.so/_backend_agg.pyd does the trick (along with passing
>> straight alphas). So far, I've seen no ill effects on any of my
>> plots, but I'm also not in a position to run the pixel-by-pixel
>> comparison matplotlib tests.
>>
>>
>> On Wed, Oct 19, 2011 at 7:26 PM, Daniel Hyams<dh...@gm...> wrote:
>>>
>>> There has to be something else in play here. I'll try to keep this
>>> short, but the summary is this: I can get the transparency to look
>>> right, but only if 1) I put "straight" alpha in image, not
>>> premultiplied, and 2) I hack agg to remove specificially every
>>> instance of the lines of code that you refer to above.
>>>
>>> Why this is, I don't know. Hopefully I'm still misusing something.
>>> However, it behaves as if the clipping of alpha in the agg library is
>>> corrupting the alpha channel. I also submit that I could have broken
>>> some other transparency capabilities of matplotlib, because I don't
>>> know what other routines use what I hacked....I did check a few
>>> transparent polygons and such though, and everything seemed to be
>>> fine.
>>>
>>> I know that the agg library has been around for quite a long time, so
>>> that also means that such a basic bug is unlikely.
>>>
>>> I've reattached the (slightly modified) script that reproduces the
>>> problem, along with a sample image that it uses. The only change to
>>> the script is right at the top, where a different image is read, a
>>> quick statement is placed to add an alpha channel if there is not
>>> already one, and I'm attempting to use premultiplied alphas. I've
>>> also attached a screenshot of the output. Notice that in this case,
>>> both "transparent" images look wrong.
>>>
>>> Now, if I 1) hack agg to remove the alpha clipping, and 2) modify the
>>> one line in the attached python script so that I use straight alpha,
>>> everything looks right. Specifically, I removed every instance of the
>>> code below from xxxx, rebuilt all of the matplotlib .so's, and
>>> specifically replaced _image.so and _backend_agg.so in my matplotlib
>>> distribution.
>>>
>>>      if(fg[order_type::A]> base_mask)     fg[order_type::A]
>>> = base_mask;
>>>        if(fg[order_type::R]> fg[order_type::A])
>>> fg[order_type::R] = fg[order_type::A];
>>>        if(fg[order_type::G]> fg[order_type::A])
>>> fg[order_type::G] = fg[order_type::A];
>>>        if(fg[order_type::B]> fg[order_type::A])
>>> fg[order_type::B] = fg[order_type::A];
>>>
>>>
>>>
>>>
>>> On Wed, Oct 19, 2011 at 2:34 PM, Daniel Hyams<dh...@gm...> wrote:
>>>>
>>>> Ah, thanks so much Michael! That explanation helps a great deal; I
>>>> was always considering things in "straight alpha" format, not even
>>>> knowing that there was alternative.
>>>>
>>>> I'll play with this tonight; I don't see any problem getting the thing
>>>> working, though, now that I know what agg expects to see...
>>>>
>>>> And yes, alpha support in the image class would be very helpful ;)
>>>>
>>>> On Wed, Oct 19, 2011 at 2:16 PM, Michael Droettboom<md...@st...>
>>>> wrote:
>>>>>
>>>>> You are right that Agg is doing the resizing here. Agg expects
>>>>> premultiplied alpha. See [1] for information about what that means.
>>>>>
>>>>> [1] http://en.wikipedia.org/wiki/Alpha_compositing
>>>>>
>>>>> After Agg interpolates the pixel values, to prevent oversaturation it
>>>>> truncates all values to be less than alpha (which makes sense if
>>>>> everything
>>>>> is assumed to be premultiplied alpha). Arguably, the bug here is that
>>>>> nearest neighbor (which doesn't have to do any blending) doesn't
>>>>> perform the
>>>>> truncation step -- then both would look "wrong".
>>>>>
>>>>> It happens in this code snippet in span_image_filter_rgba: (base_mask
>>>>> is
>>>>> 255)
>>>>>
>>>>>         if(fg[order_type::A]> base_mask)
>>>>> fg[order_type::A]
>>>>> = base_mask;
>>>>>         if(fg[order_type::R]> fg[order_type::A])
>>>>> fg[order_type::R]
>>>>> = fg[order_type::A];
>>>>>         if(fg[order_type::G]> fg[order_type::A])
>>>>> fg[order_type::G]
>>>>> = fg[order_type::A];
>>>>>         if(fg[order_type::B]> fg[order_type::A])
>>>>> fg[order_type::B]
>>>>> = fg[order_type::A];
>>>>>
>>>>> So, the solution to make a partially transparent image is to not do:
>>>>>
>>>>>   pix[:,:,3] = 127
>>>>>
>>>>> but instead, do
>>>>>
>>>>>   pix *= 0.5
>>>>>
>>>>> Of course, the real fix here is to support alpha blending properly in
>>>>> the
>>>>> image class, then the user wouldn't have to deal with such details. A
>>>>> bug
>>>>> should probably be filed in the matplotlib issue tracker for this.
>>>>>
>>>>> Mike
>>>>>
>>>>> On 10/19/2011 12:23 PM, Daniel Hyams wrote:
>>>>>
>>>>> [Sorry, I keep getting tripped up with HTML mail....resent in ascii,
>>>>> and resaved one of the attachment png's to make it smaller.]
>>>>>
>>>>>
>>>>> Example script attached (PIL required). Basically, if I impose a
>>>>> specific value into an image's alpha channel and use any interpolation
>>>>> scheme other than 'nearest', there appears gray all where the figure
>>>>> didn't have any color to begin with.  I've also attached a screenshot
>>>>> of the output of the script on my machine.
>>>>>
>>>>> Hopefully I'm doing something wrongly?
>>>>>
>>>>> I chased the problem and managed to hack in a solution that fixes the
>>>>> problem, but it's extremely inefficient...basically, in matplotlib's
>>>>> image.py, routine BboxImage.make_image, you can create two images
>>>>> there....one with no alpha channel (call it imRGB) and one with (call
>>>>> it imRGBA). Go through all of the routine, doing exactly the same
>>>>> things to both of the images *except* for the interpolation, which is
>>>>> set to 'nearest' for imRGBA. Then, rip the colors out of imRGB, the
>>>>> alpha channel off of imRGBA, and put them together....go through all
>>>>> of the routine again with this composited image, and it works. I
>>>>> know...I told you it was bad ;)
>>>>>
>>>>> The problem seems to be in the "resize" call in that routine...resize,
>>>>> which calls into C code, does not appear to handle things correctly
>>>>> when the alpha is anything other than 255's across the board. It
>>>>> might be a problem in the agg routines, but hopefully it is just maybe
>>>>> a misuse of the agg routines.
>>>>>
>>>>> The behavior seems to be backend independent as far as I could test (I
>>>>> tried with wxagg and tk backends). I am using mpl 1.0.0 on Windows if
>>>>> it matters.
>>>>>
>>>>>
>>>>> --
>>>>> Daniel Hyams
>>>>> dh...@gm...
>>>>>
>>>>>
>>>>> ------------------------------------------------------------------------------
>>>>> All the data continuously generated in your IT infrastructure contains
>>>>> a
>>>>> definitive record of customers, application performance, security
>>>>> threats, fraudulent activity and more. Splunk takes this data and makes
>>>>> sense of it. Business sense. IT sense. Common sense.
>>>>> http://p.sf.net/sfu/splunk-d2d-oct
>>>>>
>>>>> _______________________________________________
>>>>> Matplotlib-users mailing list
>>>>> Mat...@li...
>>>>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>>>>
>>>>>
>>>>>
>>>>> ------------------------------------------------------------------------------
>>>>> The demand for IT networking professionals continues to grow, and the
>>>>> demand for specialized networking skills is growing even more rapidly.
>>>>> Take a complimentary Learning@Ciosco Self-Assessment and learn
>>>>> about Cisco certifications, training, and career opportunities.
>>>>> http://p.sf.net/sfu/cisco-dev2dev
>>>>> _______________________________________________
>>>>> Matplotlib-users mailing list
>>>>> Mat...@li...
>>>>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>>>>
>>>>>
>>>>
>>>>
>>>> --
>>>> Daniel Hyams
>>>> dh...@gm...
>>>
>>>
>>> --
>>> Daniel Hyams
>>> dh...@gm...
>>>
>>
>>
>
>
> ------------------------------------------------------------------------------
> The demand for IT networking professionals continues to grow, and the
> demand for specialized networking skills is growing even more rapidly.
> Take a complimentary Learning@Cisco Self-Assessment and learn
> about Cisco certifications, training, and career opportunities.
> http://p.sf.net/sfu/cisco-dev2dev
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>
-- 
Daniel Hyams
dh...@gm...

2011年10月20日 Benjamin Root <ben...@ou...>:
> I have been doing some thinking about this problem and how it could relate
> to issues I have encountered in my work on mplot3d. The key problem I
> encounter there is that shading of the colors needs to be performed based on
> perspective (rotation of the scene and such).
>
> In many ways, this is a similar problem to wanting b&w versions of graphs.
>
> The problem I have had in addressing this is that there are two primary ways
> to specify color in mpl. First, you can specify it directly with strings or
> rgb(a) tuples. Second, you can specify it through a ScalarMappable (eg,
> colormap). This makes for two very different code paths to account for, and
> it can often be difficult to detect which path is needed.
You are facing the OO design of Mpl. I believe they cannot be solved
in mpl, only remedied partially. This is the reason why I don't want
to get entangled with mpl.
> So, what if we unify everything under ScalarMappable? The use case of
> directly specifying the color could be envisioned as the color spec being
> the data applied to an "identity colormap".
No, this would hinder further differentiation. You just gave the
example: Shading. "Uniting", i.e. seeing everything as a
ScalarMappable, would make it impossible without further "unification"
to take two parameters into account: Height + Viewing angle. But the
problem exists, as I said it stems from the Oo approach.
> Then, we extend ScalarMappable such that a chain of transforms could be
> performed on the rgba array. I envision this to be very similar to how the
> projections framework operates. We can then have a set of stock transforms
> that users could select at will (such as color to b&w).
Exactly. But this is a functional approach. Isn't it?
> We can use the lessons and successes of the projections framework to guide
> our design here.
I strongly believe, although I don't know the projections framework,
that is has functional ideas strongly expressed in the way it works.
Is that true?
I just had a walk and tapped in my thoughts into my notebook device
while walking, this is some way of meditation. So here they come (The
postscripta are now-thoughts):
1. There are two different ways of specifying a colour: Either a
thing is simply red or you need something that spits out a colour when
the specification is given. Uniting both under the
spit-out-with-specification constraint hinders further
differentiation. P.S.: In fact, it would be better if the "red" spits
out a colour when called with *nothing*. But hey, this is functional.
2. In fact I believe the Oo approach does not work well for frame
generation (and plotting is a frame generation). This is why I coded
moviemaker using functional paradigm. It just didn't work out the Oo
way - just because you needed a new class for each type of plotting
directive; you have to pass everything through. The functional
approach solved my headaches - Classes are too rigid for plotting.
3. I don't want to "pollute" my mind with the Oo ideas of mpl; rather
I would like to learn to avoid the mistakes made here.
4. Your idea is a functional one.
5. In functional programming, each object (i.e. each function) has
one entry point, in Python this is implemented by the __call__()
method.
-
I can give further details but it doesn't matter here, since we're
talking to give mpl grayscale and filter properties, right. I will
try to dig out my old private ideas I sent you some monthes ago - I
believe I already found the solution for our matter here. (i.e.
avoiding rcParams and stuff.)
I believe the one and only solution would, if thought thru completely,
unveil that we need functional approach to get better results. In
matplotlib, each type of structure is an object of a class, which is
not callable mostly, and this means, you have to formulate each
alteration explicitly. It's not easy in this frame of thinking to get
the flexibility we in fact want - this might be the reason why mpl is
so monolithic with all its different kind of Artist objects ... with
the counterparts in the Renderer objects and so on.
It's hard to keep a common style when differentiating things in
plotting using the Oo approach.
This is just what we see in fact. We feel that things are somehow
united and should work better together and be more exchangable, but
they aren't, because it has never been formulated in terms of code, of
a common base class Function, as I would do it (and did it in
moviemaker). Addendum: It has been attempted with the Artist class,
with the .render() method. I cannot judge how well it worked out, but
to my understanding, it didn't foster plug-in architecture much. ADD:
I think this is because .render() happens on high-level, so it's not
attractive for data-munging objects. The functional approach is
attractive there.
This can be understood as some advertisement for plotlayers, but
believe me, I have coded not a single line of it so far. There exists
moviemaker3, though, and that one works fine. What's missing are
things like polygons and fast drawing routines for scaled pixmaps.
I'll not go into details here.
For me it's hard to stay focussed on mpl here.
I will try to dig out that emails.
Friedrich

Thanks for looking into this deeper.
Agg requires image buffers to be premultiplied, as described in the 
third bullet point here. (It's not exactly clear, to say the least, but 
that's what I take it to mean, and also from reading the code).
http://www.antigrain.com/news/release_notes/v22.agdoc.html
The bug is that in _image.cpp the input buffers are not declared as 
premultiplied in _image.cpp. Arguably it is a bug that agg doesn't 
reject filtering unmultiplied images, since the note states that the 
assumption is that they are premultiplied by the time they get to the 
filters.
I have attached a patch that fixes this. Would you mind testing it and 
let me know how it works for you?
On 10/20/2011 10:29 PM, Daniel Hyams wrote:
> I've looked all over the place through both the Python and C code, and
> I don't see any premultiplication of alphas at any stage before the
> pixels are passed off to agg, and neither can I find any place where
> the alphas are "unmultiplied" on the way back from agg to the backend
> for rendering.
matplotlib's support for alpha blending of images is basically by 
accident, so it's not surprising the details aren't right.
I think it is a bug that after reading images in we don't premultiply 
them before sending them to Agg. That bug has existed for a long time 
in matplotlib because no one is really using alpha images a great deal. 
(Masked images, yes, but that implies alpha is strictly 0 or 255 and 
thus these issues don't come into play.)
Unmultiplying is not always necessary. Many of the GUI backends also 
expect premultiplied alpha (Qt for example). However, there is 
certainly a bug in writing PNG files (where the file format specifies 
unmultiplied).
> It's very possible that I missed it, but I would have to miss it in
> two places (premultiply and the unmultiply). It looks to me like the
> output from agg ends up getting passed on directly to the renderer,
> which as far as I know, just uses straight alpha. The WxAgg renderer,
> for example, just creates a wx.Bitmap out of the pixels and blits it.
> Which means that any image going through agg's filters will not be
> correct if it has any pixels with alpha != 0 or != 255.
>
> [Using PIL images because they are simple to talk about...but the PIL
> image could alternatively be an image.py image]
>
> As far as I can tell, the image pixels current go through a pipeline
> like the following:
>
> [1] PIL Image -> _image image -> agg operations -> modified and/or
> resized _image image -> renderer
>
> If agg expects premultiplied alpha, the procedure should look something like:
>
> [2] PIL Image -> _image image -> premultiply alphas ->agg options ->
> unmultiply alphas -> modified and/or resized _image image -> renderer
>
> I personally don't like pipeline [2] because picture detail is lost in
> the "unmultiply alphas" stage. Better to use straight alpha all the
> way through.
I think what needed is:
[3] PIL Image (or _png.cpp) -> premultiply alphas -> _image image -> agg options ->
-> modified and/or resized _image image -> renderer -> (unmultiply alphas)? -> GUI library
That is -- all image data should be kept premultiplied internally in all 
buffers for efficiency and because this is what Agg is designed for.
Can you explain what you mean by "picture detail is lost in the 
unmultiply alphas stage". There is the usual problem that by 
premultiplying you lose any color data where alpha = 0 (and you lose 
resolution everywhere else, but not resolution you can actually see 
after compositing).
> So long as matplotlib is using only a subset of agg algorithms that
> work no matter whether the alphas are premultiplied or not, I would
> think that the most reasonable route was the one that I took; to
> always pass straight alphas (sticking with pipeline [1]), and modify
> the agg source slightly to fit matplotlib's approach (i.e., remove the
> clipping there).
I'd be really wary of modifying agg like this. Those things become hard 
to maintain. I think this instead a bug in matplotlib and should be 
fixed there.
I've put an issue in the issue tracker here:
https://github.com/matplotlib/matplotlib/issues/545
Cheers,
Mike
> I hope that I'm not way off base (I have a sneaking feeling that I am
> :O ), and hope this helps. I've verified on both Linux and Windows
> that removing the alpha-clip lines from agg_span_image_filter_rgba.h,
> rebuilding matplotlib, and replacing _image.so/_image.pyd and
> _backend_agg.so/_backend_agg.pyd does the trick (along with passing
> straight alphas). So far, I've seen no ill effects on any of my
> plots, but I'm also not in a position to run the pixel-by-pixel
> comparison matplotlib tests.
>
>
> On Wed, Oct 19, 2011 at 7:26 PM, Daniel Hyams<dh...@gm...> wrote:
>> There has to be something else in play here. I'll try to keep this
>> short, but the summary is this: I can get the transparency to look
>> right, but only if 1) I put "straight" alpha in image, not
>> premultiplied, and 2) I hack agg to remove specificially every
>> instance of the lines of code that you refer to above.
>>
>> Why this is, I don't know. Hopefully I'm still misusing something.
>> However, it behaves as if the clipping of alpha in the agg library is
>> corrupting the alpha channel. I also submit that I could have broken
>> some other transparency capabilities of matplotlib, because I don't
>> know what other routines use what I hacked....I did check a few
>> transparent polygons and such though, and everything seemed to be
>> fine.
>>
>> I know that the agg library has been around for quite a long time, so
>> that also means that such a basic bug is unlikely.
>>
>> I've reattached the (slightly modified) script that reproduces the
>> problem, along with a sample image that it uses. The only change to
>> the script is right at the top, where a different image is read, a
>> quick statement is placed to add an alpha channel if there is not
>> already one, and I'm attempting to use premultiplied alphas. I've
>> also attached a screenshot of the output. Notice that in this case,
>> both "transparent" images look wrong.
>>
>> Now, if I 1) hack agg to remove the alpha clipping, and 2) modify the
>> one line in the attached python script so that I use straight alpha,
>> everything looks right. Specifically, I removed every instance of the
>> code below from xxxx, rebuilt all of the matplotlib .so's, and
>> specifically replaced _image.so and _backend_agg.so in my matplotlib
>> distribution.
>>
>> if(fg[order_type::A]> base_mask) fg[order_type::A]
>> = base_mask;
>> if(fg[order_type::R]> fg[order_type::A])
>> fg[order_type::R] = fg[order_type::A];
>> if(fg[order_type::G]> fg[order_type::A])
>> fg[order_type::G] = fg[order_type::A];
>> if(fg[order_type::B]> fg[order_type::A])
>> fg[order_type::B] = fg[order_type::A];
>>
>>
>>
>>
>> On Wed, Oct 19, 2011 at 2:34 PM, Daniel Hyams<dh...@gm...> wrote:
>>> Ah, thanks so much Michael! That explanation helps a great deal; I
>>> was always considering things in "straight alpha" format, not even
>>> knowing that there was alternative.
>>>
>>> I'll play with this tonight; I don't see any problem getting the thing
>>> working, though, now that I know what agg expects to see...
>>>
>>> And yes, alpha support in the image class would be very helpful ;)
>>>
>>> On Wed, Oct 19, 2011 at 2:16 PM, Michael Droettboom<md...@st...> wrote:
>>>> You are right that Agg is doing the resizing here. Agg expects
>>>> premultiplied alpha. See [1] for information about what that means.
>>>>
>>>> [1] http://en.wikipedia.org/wiki/Alpha_compositing
>>>>
>>>> After Agg interpolates the pixel values, to prevent oversaturation it
>>>> truncates all values to be less than alpha (which makes sense if everything
>>>> is assumed to be premultiplied alpha). Arguably, the bug here is that
>>>> nearest neighbor (which doesn't have to do any blending) doesn't perform the
>>>> truncation step -- then both would look "wrong".
>>>>
>>>> It happens in this code snippet in span_image_filter_rgba: (base_mask is
>>>> 255)
>>>>
>>>> if(fg[order_type::A]> base_mask) fg[order_type::A]
>>>> = base_mask;
>>>> if(fg[order_type::R]> fg[order_type::A]) fg[order_type::R]
>>>> = fg[order_type::A];
>>>> if(fg[order_type::G]> fg[order_type::A]) fg[order_type::G]
>>>> = fg[order_type::A];
>>>> if(fg[order_type::B]> fg[order_type::A]) fg[order_type::B]
>>>> = fg[order_type::A];
>>>>
>>>> So, the solution to make a partially transparent image is to not do:
>>>>
>>>> pix[:,:,3] = 127
>>>>
>>>> but instead, do
>>>>
>>>> pix *= 0.5
>>>>
>>>> Of course, the real fix here is to support alpha blending properly in the
>>>> image class, then the user wouldn't have to deal with such details. A bug
>>>> should probably be filed in the matplotlib issue tracker for this.
>>>>
>>>> Mike
>>>>
>>>> On 10/19/2011 12:23 PM, Daniel Hyams wrote:
>>>>
>>>> [Sorry, I keep getting tripped up with HTML mail....resent in ascii,
>>>> and resaved one of the attachment png's to make it smaller.]
>>>>
>>>>
>>>> Example script attached (PIL required). Basically, if I impose a
>>>> specific value into an image's alpha channel and use any interpolation
>>>> scheme other than 'nearest', there appears gray all where the figure
>>>> didn't have any color to begin with. I've also attached a screenshot
>>>> of the output of the script on my machine.
>>>>
>>>> Hopefully I'm doing something wrongly?
>>>>
>>>> I chased the problem and managed to hack in a solution that fixes the
>>>> problem, but it's extremely inefficient...basically, in matplotlib's
>>>> image.py, routine BboxImage.make_image, you can create two images
>>>> there....one with no alpha channel (call it imRGB) and one with (call
>>>> it imRGBA). Go through all of the routine, doing exactly the same
>>>> things to both of the images *except* for the interpolation, which is
>>>> set to 'nearest' for imRGBA. Then, rip the colors out of imRGB, the
>>>> alpha channel off of imRGBA, and put them together....go through all
>>>> of the routine again with this composited image, and it works. I
>>>> know...I told you it was bad ;)
>>>>
>>>> The problem seems to be in the "resize" call in that routine...resize,
>>>> which calls into C code, does not appear to handle things correctly
>>>> when the alpha is anything other than 255's across the board. It
>>>> might be a problem in the agg routines, but hopefully it is just maybe
>>>> a misuse of the agg routines.
>>>>
>>>> The behavior seems to be backend independent as far as I could test (I
>>>> tried with wxagg and tk backends). I am using mpl 1.0.0 on Windows if
>>>> it matters.
>>>>
>>>>
>>>> --
>>>> Daniel Hyams
>>>> dh...@gm...
>>>>
>>>> ------------------------------------------------------------------------------
>>>> All the data continuously generated in your IT infrastructure contains a
>>>> definitive record of customers, application performance, security
>>>> threats, fraudulent activity and more. Splunk takes this data and makes
>>>> sense of it. Business sense. IT sense. Common sense.
>>>> http://p.sf.net/sfu/splunk-d2d-oct
>>>>
>>>> _______________________________________________
>>>> Matplotlib-users mailing list
>>>> Mat...@li...
>>>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>>>
>>>>
>>>> ------------------------------------------------------------------------------
>>>> The demand for IT networking professionals continues to grow, and the
>>>> demand for specialized networking skills is growing even more rapidly.
>>>> Take a complimentary Learning@Ciosco Self-Assessment and learn
>>>> about Cisco certifications, training, and career opportunities.
>>>> http://p.sf.net/sfu/cisco-dev2dev
>>>> _______________________________________________
>>>> Matplotlib-users mailing list
>>>> Mat...@li...
>>>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>>>
>>>>
>>>
>>>
>>> --
>>> Daniel Hyams
>>> dh...@gm...
>>
>>
>> --
>> Daniel Hyams
>> dh...@gm...
>>
>
>

Dear list,
I am making a 3d plot of a set of functions one after the other.
Actually I do this interactively, but what matter should only be that
I plot, clear the axis, plot again and so on.
I want to display a colorbar for each of those plots. The first thing
I tried was calling
 fig.colorbar(surf, shrink=0.5, aspect=10)
each time. This produced a new colorbar each time I called it. So
thats not working, but thats ok.
My question is now, how do I update the colorbar? I tried several methods:
 #self.cbShape.update_bruteforce(surf)
 self.cbShape.update_normal(surf)
 self.cbShape.changed()
 self.cbShape.update_ticks()
but none of them updated the colorbar. So the plot clearly used colors
from [0, 0.2] but the colorbar always stays at [0, 0.05] (which are
the values from the first time I drew).
How to correctly update the colorbar?
-Holger

Indeed, with that link, I was able to make it work.
Thanks a lot,
Holger
On Thu, Oct 20, 2011 at 11:30, Jean-Louis Durrieu <jea...@du...> wrote:
> Hi Holger,
>
> On Oct 20, 2011, at 11:21 AM, Holger Brandsmeier wrote:
>> I would like to have two subplots in a figure one below each other (so
>> arranged in a 2x1 matrix). I want the upper subplot to have a very
>> small height compared to the lower one, say 10 times smaller than the
>> other one.
>>
>> How can I do that with matplotlib? I didn't find that with subplot()
>> command. I also found the add_subplot() function with many parameters,
>> but I didn't see which of those parameter would help me.
>
> Is this what you need?
> http://matplotlib.sourceforge.net/mpl_toolkits/axes_grid/users/overview.html#scatter-hist-py-with-axesdivider
>
> Best regards,
>
> Jean-Louis
>
>>
>> Thanks for any advice,
>> Holger
>>
>> --
>> Holger Brandsmeier, SAM, ETH Zürich
>> http://www.sam.math.ethz.ch/people/bholger
>>
>> ------------------------------------------------------------------------------
>> The demand for IT networking professionals continues to grow, and the
>> demand for specialized networking skills is growing even more rapidly.
>> Take a complimentary Learning@Ciosco Self-Assessment and learn
>> about Cisco certifications, training, and career opportunities.
>> http://p.sf.net/sfu/cisco-dev2dev
>> _______________________________________________
>> Matplotlib-users mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>
>
>
-- 
Holger Brandsmeier, SAM, ETH Zürich
http://www.sam.math.ethz.ch/people/bholger

On 10/20/2011 05:17 PM, John Gu wrote:
> Thanks. Is there a place where these sorts of issues can be submitted
> for review / fixes? I'm totally willing to take a look at possible
> solutions if someone points me in the right direction. Thanks.
John,
Development takes place via github, so you can submit issues, clone and 
make pull requests, etc.:
https://github.com/matplotlib/matplotlib
Eric

Thanks. Is there a place where these sorts of issues can be submitted for
review / fixes? I'm totally willing to take a look at possible solutions if
someone points me in the right direction. Thanks.
On Thu, Oct 20, 2011 at 10:52 PM, Eric Firing <ef...@ha...> wrote:
> On 10/20/2011 03:47 PM, John Gu wrote:
> > Hello,
> >
> > I'm using version 1.0.1 of matplotlib on a linux machine. uname -a
> > returns the following: Linux jgulinux 2.6.35.14-95.fc14.x86_64 #1 SMP
> > Tue Aug 16 21:01:58 UTC 2011 x86_64 x86_64 x86_64 GNU/Linux.
> >
> > [jgu@jgulinux ~/.matplotlib]$ cat matplotlibrc
> > backend : TkAgg
> > interactive : True
> >
> > Program that reproduces the problem (run in ipython):
> >
> > x = arange(1000)
> > y = x
> >
> > f = figure(1)
> > for ia in xrange(5):
> > for ib in xrange(5):
> > ax = subplot(5,5,ia*5+ib+1, navigate=True)
> > ax.plot(x,y)
> >
> > If we now tried to interactively navigate through any single subplot of
> > the figure, it is extremely slow. Not sure if this is a configuration
> > error on my side?
>
> No, and it is not backend-dependent, either. And it does not even
> require that all of the plots have data--it is the same if only the last
> subplot has data. A little experimentation indicates that all the ticks
> and tick labels on the other subplots are slowing down the interaction
> with the one subplot being navigated.
>
> ax.tick_params(labelbottom=False, labelleft=False)
> ax.tick_params(left=False, right=False, top=False, bottom=False)
>
> Insert the above to turn off ticks and tick labels, and it speeds up
> nicely. That's not a solution, I realize.
>
> I knew that slowness in the tick and tick label generation was a major
> bottleneck in subplot creation, but I did not realize that it had such
> an effect on interactive manipulation of a single subplot.
>
> Eric
>
> >
> > John
>
>
> ------------------------------------------------------------------------------
> The demand for IT networking professionals continues to grow, and the
> demand for specialized networking skills is growing even more rapidly.
> Take a complimentary Learning@Cisco Self-Assessment and learn
> about Cisco certifications, training, and career opportunities.
> http://p.sf.net/sfu/cisco-dev2dev
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

On 10/20/2011 03:47 PM, John Gu wrote:
> Hello,
>
> I'm using version 1.0.1 of matplotlib on a linux machine. uname -a
> returns the following: Linux jgulinux 2.6.35.14-95.fc14.x86_64 #1 SMP
> Tue Aug 16 21:01:58 UTC 2011 x86_64 x86_64 x86_64 GNU/Linux.
>
> [jgu@jgulinux ~/.matplotlib]$ cat matplotlibrc
> backend : TkAgg
> interactive : True
>
> Program that reproduces the problem (run in ipython):
>
> x = arange(1000)
> y = x
>
> f = figure(1)
> for ia in xrange(5):
> for ib in xrange(5):
> ax = subplot(5,5,ia*5+ib+1, navigate=True)
> ax.plot(x,y)
>
> If we now tried to interactively navigate through any single subplot of
> the figure, it is extremely slow. Not sure if this is a configuration
> error on my side?
No, and it is not backend-dependent, either. And it does not even 
require that all of the plots have data--it is the same if only the last 
subplot has data. A little experimentation indicates that all the ticks 
and tick labels on the other subplots are slowing down the interaction 
with the one subplot being navigated.
 ax.tick_params(labelbottom=False, labelleft=False)
 ax.tick_params(left=False, right=False, top=False, bottom=False)
Insert the above to turn off ticks and tick labels, and it speeds up 
nicely. That's not a solution, I realize.
I knew that slowness in the tick and tick label generation was a major 
bottleneck in subplot creation, but I did not realize that it had such 
an effect on interactive manipulation of a single subplot.
Eric
>
> John

I've looked all over the place through both the Python and C code, and
I don't see any premultiplication of alphas at any stage before the
pixels are passed off to agg, and neither can I find any place where
the alphas are "unmultiplied" on the way back from agg to the backend
for rendering.
It's very possible that I missed it, but I would have to miss it in
two places (premultiply and the unmultiply). It looks to me like the
output from agg ends up getting passed on directly to the renderer,
which as far as I know, just uses straight alpha. The WxAgg renderer,
for example, just creates a wx.Bitmap out of the pixels and blits it.
Which means that any image going through agg's filters will not be
correct if it has any pixels with alpha != 0 or != 255.
[Using PIL images because they are simple to talk about...but the PIL
image could alternatively be an image.py image]
As far as I can tell, the image pixels current go through a pipeline
like the following:
[1] PIL Image -> _image image -> agg operations -> modified and/or
resized _image image -> renderer
If agg expects premultiplied alpha, the procedure should look something like:
[2] PIL Image -> _image image -> premultiply alphas ->agg options ->
unmultiply alphas -> modified and/or resized _image image -> renderer
I personally don't like pipeline [2] because picture detail is lost in
the "unmultiply alphas" stage. Better to use straight alpha all the
way through.
So long as matplotlib is using only a subset of agg algorithms that
work no matter whether the alphas are premultiplied or not, I would
think that the most reasonable route was the one that I took; to
always pass straight alphas (sticking with pipeline [1]), and modify
the agg source slightly to fit matplotlib's approach (i.e., remove the
clipping there).
I hope that I'm not way off base (I have a sneaking feeling that I am
:O ), and hope this helps. I've verified on both Linux and Windows
that removing the alpha-clip lines from agg_span_image_filter_rgba.h,
rebuilding matplotlib, and replacing _image.so/_image.pyd and
_backend_agg.so/_backend_agg.pyd does the trick (along with passing
straight alphas). So far, I've seen no ill effects on any of my
plots, but I'm also not in a position to run the pixel-by-pixel
comparison matplotlib tests.
On Wed, Oct 19, 2011 at 7:26 PM, Daniel Hyams <dh...@gm...> wrote:
> There has to be something else in play here. I'll try to keep this
> short, but the summary is this: I can get the transparency to look
> right, but only if 1) I put "straight" alpha in image, not
> premultiplied, and 2) I hack agg to remove specificially every
> instance of the lines of code that you refer to above.
>
> Why this is, I don't know. Hopefully I'm still misusing something.
> However, it behaves as if the clipping of alpha in the agg library is
> corrupting the alpha channel. I also submit that I could have broken
> some other transparency capabilities of matplotlib, because I don't
> know what other routines use what I hacked....I did check a few
> transparent polygons and such though, and everything seemed to be
> fine.
>
> I know that the agg library has been around for quite a long time, so
> that also means that such a basic bug is unlikely.
>
> I've reattached the (slightly modified) script that reproduces the
> problem, along with a sample image that it uses. The only change to
> the script is right at the top, where a different image is read, a
> quick statement is placed to add an alpha channel if there is not
> already one, and I'm attempting to use premultiplied alphas. I've
> also attached a screenshot of the output. Notice that in this case,
> both "transparent" images look wrong.
>
> Now, if I 1) hack agg to remove the alpha clipping, and 2) modify the
> one line in the attached python script so that I use straight alpha,
> everything looks right. Specifically, I removed every instance of the
> code below from xxxx, rebuilt all of the matplotlib .so's, and
> specifically replaced _image.so and _backend_agg.so in my matplotlib
> distribution.
>
>      if(fg[order_type::A] > base_mask)     fg[order_type::A]
> = base_mask;
>        if(fg[order_type::R] > fg[order_type::A])
> fg[order_type::R] = fg[order_type::A];
>        if(fg[order_type::G] > fg[order_type::A])
> fg[order_type::G] = fg[order_type::A];
>        if(fg[order_type::B] > fg[order_type::A])
> fg[order_type::B] = fg[order_type::A];
>
>
>
>
> On Wed, Oct 19, 2011 at 2:34 PM, Daniel Hyams <dh...@gm...> wrote:
>>
>> Ah, thanks so much Michael! That explanation helps a great deal; I
>> was always considering things in "straight alpha" format, not even
>> knowing that there was alternative.
>>
>> I'll play with this tonight; I don't see any problem getting the thing
>> working, though, now that I know what agg expects to see...
>>
>> And yes, alpha support in the image class would be very helpful ;)
>>
>> On Wed, Oct 19, 2011 at 2:16 PM, Michael Droettboom <md...@st...> wrote:
>> > You are right that Agg is doing the resizing here. Agg expects
>> > premultiplied alpha. See [1] for information about what that means.
>> >
>> > [1] http://en.wikipedia.org/wiki/Alpha_compositing
>> >
>> > After Agg interpolates the pixel values, to prevent oversaturation it
>> > truncates all values to be less than alpha (which makes sense if everything
>> > is assumed to be premultiplied alpha). Arguably, the bug here is that
>> > nearest neighbor (which doesn't have to do any blending) doesn't perform the
>> > truncation step -- then both would look "wrong".
>> >
>> > It happens in this code snippet in span_image_filter_rgba: (base_mask is
>> > 255)
>> >
>> >         if(fg[order_type::A] > base_mask)     fg[order_type::A]
>> > = base_mask;
>> >         if(fg[order_type::R] > fg[order_type::A]) fg[order_type::R]
>> > = fg[order_type::A];
>> >         if(fg[order_type::G] > fg[order_type::A]) fg[order_type::G]
>> > = fg[order_type::A];
>> >         if(fg[order_type::B] > fg[order_type::A]) fg[order_type::B]
>> > = fg[order_type::A];
>> >
>> > So, the solution to make a partially transparent image is to not do:
>> >
>> >   pix[:,:,3] = 127
>> >
>> > but instead, do
>> >
>> >   pix *= 0.5
>> >
>> > Of course, the real fix here is to support alpha blending properly in the
>> > image class, then the user wouldn't have to deal with such details. A bug
>> > should probably be filed in the matplotlib issue tracker for this.
>> >
>> > Mike
>> >
>> > On 10/19/2011 12:23 PM, Daniel Hyams wrote:
>> >
>> > [Sorry, I keep getting tripped up with HTML mail....resent in ascii,
>> > and resaved one of the attachment png's to make it smaller.]
>> >
>> >
>> > Example script attached (PIL required). Basically, if I impose a
>> > specific value into an image's alpha channel and use any interpolation
>> > scheme other than 'nearest', there appears gray all where the figure
>> > didn't have any color to begin with.  I've also attached a screenshot
>> > of the output of the script on my machine.
>> >
>> > Hopefully I'm doing something wrongly?
>> >
>> > I chased the problem and managed to hack in a solution that fixes the
>> > problem, but it's extremely inefficient...basically, in matplotlib's
>> > image.py, routine BboxImage.make_image, you can create two images
>> > there....one with no alpha channel (call it imRGB) and one with (call
>> > it imRGBA). Go through all of the routine, doing exactly the same
>> > things to both of the images *except* for the interpolation, which is
>> > set to 'nearest' for imRGBA. Then, rip the colors out of imRGB, the
>> > alpha channel off of imRGBA, and put them together....go through all
>> > of the routine again with this composited image, and it works. I
>> > know...I told you it was bad ;)
>> >
>> > The problem seems to be in the "resize" call in that routine...resize,
>> > which calls into C code, does not appear to handle things correctly
>> > when the alpha is anything other than 255's across the board. It
>> > might be a problem in the agg routines, but hopefully it is just maybe
>> > a misuse of the agg routines.
>> >
>> > The behavior seems to be backend independent as far as I could test (I
>> > tried with wxagg and tk backends). I am using mpl 1.0.0 on Windows if
>> > it matters.
>> >
>> >
>> > --
>> > Daniel Hyams
>> > dh...@gm...
>> >
>> > ------------------------------------------------------------------------------
>> > All the data continuously generated in your IT infrastructure contains a
>> > definitive record of customers, application performance, security
>> > threats, fraudulent activity and more. Splunk takes this data and makes
>> > sense of it. Business sense. IT sense. Common sense.
>> > http://p.sf.net/sfu/splunk-d2d-oct
>> >
>> > _______________________________________________
>> > Matplotlib-users mailing list
>> > Mat...@li...
>> > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>> >
>> >
>> > ------------------------------------------------------------------------------
>> > The demand for IT networking professionals continues to grow, and the
>> > demand for specialized networking skills is growing even more rapidly.
>> > Take a complimentary Learning@Ciosco Self-Assessment and learn
>> > about Cisco certifications, training, and career opportunities.
>> > http://p.sf.net/sfu/cisco-dev2dev
>> > _______________________________________________
>> > Matplotlib-users mailing list
>> > Mat...@li...
>> > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>> >
>> >
>>
>>
>>
>> --
>> Daniel Hyams
>> dh...@gm...
>
>
>
> --
> Daniel Hyams
> dh...@gm...
>
-- 
Daniel Hyams
dh...@gm...

Hello,
I'm using version 1.0.1 of matplotlib on a linux machine. uname -a returns
the following: Linux jgulinux 2.6.35.14-95.fc14.x86_64 #1 SMP Tue Aug 16
21:01:58 UTC 2011 x86_64 x86_64 x86_64 GNU/Linux.
[jgu@jgulinux ~/.matplotlib]$ cat matplotlibrc
backend : TkAgg
interactive : True
Program that reproduces the problem (run in ipython):
x = arange(1000)
y = x
f = figure(1)
for ia in xrange(5):
 for ib in xrange(5):
 ax = subplot(5,5,ia*5+ib+1, navigate=True)
 ax.plot(x,y)
If we now tried to interactively navigate through any single subplot of the
figure, it is extremely slow. Not sure if this is a configuration error on
my side?
John