matplotlib-users Mailing List for matplotlib (Page 3)

"Deen Sethanandha" <khu...@gm...> writes:
> I use matplotlib as part of my Trac plugin. I got this error when I try
> to access the web site that use my plugin. [...]
>
> File "/usr/lib/python2.5/site-packages/matplotlib/pylab.py", line 876, in
> figure
> File
> "/usr/lib/python2.5/site-packages/matplotlib/backends/backend_tkagg.py",
> line 88, in new_figure_manager
> File "lib-tk/Tkinter.py", line 1639, in __init__
Your plugin is importing pylab, which automatically imports the TkAgg
backend based on your .matplotlibrc setting, and this makes no sense 
in a non-interactive environment. The quick way to make this work is 
to replace "import pylab" by the following lines:
import matplotlib
matplotlib.use('Agg')
import pylab
See also: examples/webapp_demo.py.
-- 
Jouni K. Seppänen
http://www.iki.fi/jks

Hi,
 I use matplotlib as part of my Trac plugin. I got this error when I try
to access the web site that use my plugin. Please see the trace back below
File "/u/bhuricha/lib/python2.5/site-packages/Trac-
0.11dev_r5933-py2.5.egg/trac/web/main.py", line 381, in dispatch_request
File "/u/bhuricha/lib/python2.5/site-packages/Trac-
0.11dev_r5933-py2.5.egg/trac/web/main.py", line 191, in dispatch
File
"/u/bhuricha/source/trac-hacks/tracmetrixplugin/0.11/tracmetrixplugin/web_ui.py",
line 119, in process_request
File
"/u/bhuricha/source/trac-hacks/tracmetrixplugin/0.11/tracmetrixplugin/web_ui.py",
line 193, in _render_view
File
"/u/bhuricha/source/trac-hacks/tracmetrixplugin/0.11/tracmetrixplugin/model.py",
line 452, in get_daily_backlog_chart
File "/usr/lib/python2.5/site-packages/matplotlib/pylab.py", line 876, in
figure
File
"/usr/lib/python2.5/site-packages/matplotlib/backends/backend_tkagg.py",
line 88, in new_figure_manager
File "lib-tk/Tkinter.py", line 1639, in __init__
Local variables:
Name Value
baseName 'trac.cgi'
className 'Tk'
ext '.cgi'
interactive 0
os <module 'os' from '/usr/lib/python2.5/os.pyc'>
screenName None
self None
sync 0
sys <module 'sys' (built-in)>
use None
useTk 1
 I am not sure what's wrong. I was able to get this to work in the
different environment. Can anyone tell me what might be wrong?
Thanks,
Deen

"John Hunter" <jd...@gm...> writes:
> On 8/24/07, Christopher Barker <Chr...@no...> wrote:
>
>> This may not be what it seems. The native coordinate system for
>> PostScript is in points, which are 1/72 if an inch, so it's common to
>> force that as a dpi. [...]
>
> Yes, this is exactly right and the reason we do it this way. We
> support fractional points so indeed you have higher resolutions. Off
> the top of my head, I am not sure what is going on with the image
> resolution problem....
I just tried with current svn, and the following script produces two
results that have visibly different resolutions:
#!/usr/bin/python
from pylab import *
foo = rand(10,10)
imshow(foo)
savefig('foo10.ps', dpi=10)
savefig('foo100.ps', dpi=100)
Perhaps the original poster could show a bit of code where the scaling
fails?
(I'm not sure if figimage is doing the right thing, though...)
-- 
Jouni K. Seppänen
http://www.iki.fi/jks

On 8/24/07, David Tremouilles <dav...@gm...> wrote:
> OK I see... nothing straightforward...
>
> Best way for me is maybe to implement such a system myself:
> The system would collect the information to be saved by kind of
> introspection of the figure.
> I'm planning to save data and plot properties in an hdf5 file. Kind of
> inverted process will be used to restore the figure.
> Of course I will not cover the whole possible figure case but only
> what I'm currently using for my work.
> If somebody did similar work and is eager to share or
> if somebody have any suggestion please let me know.
A huge +1 for this approach. Pickle is NOT meant to be a persistent,
on-disk file format, but rather a way to serialize the *current* state
of an object in memory. Emphasis on current: if you unpickle an old
pickle in an environment where the class layout of your object (or any
object the parent holds a reference to) has changed, the unpickling
fails, completely and irrecoverably.
As someone who has already had to write pickle loader functions to
salvage old pickles (because computing them had been very expensive),
I've learned my lesson. Pickle works well as a way to quickly
dump/load data that is either made up of simple python types *only*
(since they don't change often) or for objects that you have good
reason to expect won't be changing their API while you care about the
pickles.
But pickling is NOT a 'data format', and using it as such will
inevitably lead to much pain and suffering. HDF5 *is* a data format.
In our project we precisely went to hdf5 instead of pickling.
Cheers,
f

On 8/24/07, Christopher Barker <Chr...@no...> wrote:
> This may not be what it seems. The native coordinate system for
> PostScript is in points, which are 1/72 if an inch, so it's common to
> force that as a dpi. Postscript supports fractional (is it floating
> point or fixed -- I'm not sure) points, however, so you can define
> things in higher resolution, and I'm pretty sure you can imbed an
> arbitrary dpi image in a a PostScript file, using the 72dpi to positions
> the image.
Yes, this is exactly right and the reason we do it this way. We
support fractional points so indeed you have higher resolutions. Off
the top of my head, I am not sure what is going on with the image
resolution problem....
JDH

Petr Danecek wrote:
> Hi,
> I'd like to open a high-quality image (600dpi) in matplotlib, add some
> plots and save it as a postscript file. 
> It seems that whatever I do, the input image gets scaled down
> :-(
I'm sorry I don't know enough about MPL's handling of images to help, but...
> Looking in the code of matplotlib, there is a constant of 72dpi
> hardwired everywhere. For instance, backend_ps.py there is the line
> 	self.figure.dpi.set(72) # ignore the dpi kwarg
This may not be what it seems. The native coordinate system for 
PostScript is in points, which are 1/72 if an inch, so it's common to 
force that as a dpi. Postscript supports fractional (is it floating 
point or fixed -- I'm not sure) points, however, so you can define 
things in higher resolution, and I'm pretty sure you can imbed an 
arbitrary dpi image in a a PostScript file, using the 72dpi to positions 
the image.
The problem comes if the code, in addition to using 72dpi, also assumes 
integer coordinates, then you can't get better accuracy that 72dpi, 
which is not very good, and is really bad if someone scales it up later. 
wxWidgets addresses this by hard-coding 720dpi, rather than 72, and 
dividing by ten when writing the postscript -- still a bit of kludge. 
I'm not sure what MPL does.
-Chris
-- 
Christopher Barker, Ph.D.
Oceanographer
Emergency Response Division
NOAA/NOS/OR&R (206) 526-6959 voice
7600 Sand Point Way NE (206) 526-6329 fax
Seattle, WA 98115 (206) 526-6317 main reception
Chr...@no...

On 8/24/07, Fred Ludlow <rf...@ca...> wrote:
> # Get min and max from current axes
> y_min, y_max = gca().get_ylim()
>
> # Set the other way round for current axes
> gca().set_ylim(y_max, y_min)
>
> # Redraw
> draw()
>
>
> Emre Ayd?n wrote:
> > hi. i'm a new user of matplotlib. i've searched through the arhieves
> > of the mailing list but couldn't find a quick solution. i simply need
> > to reverse the y axis of my plots. for example in a range of 0-10, 0
> > must seem at the top of the plot where 10 is lowest, near to the x
> > axis. is there a quick solution for this. (x axis must stay as it is)
> >
> > thanx...
> >
>
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems? Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >> http://get.splunk.com/
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
thnx alot! that solved it...
-- 
EMRE AYDIN
emr...@gm...

# Get min and max from current axes
y_min, y_max = gca().get_ylim()
# Set the other way round for current axes
gca().set_ylim(y_max, y_min)
# Redraw
draw()
Emre Ayd?n wrote:
> hi. i'm a new user of matplotlib. i've searched through the arhieves
> of the mailing list but couldn't find a quick solution. i simply need
> to reverse the y axis of my plots. for example in a range of 0-10, 0
> must seem at the top of the plot where 10 is lowest, near to the x
> axis. is there a quick solution for this. (x axis must stay as it is)
> 
> thanx...
> 

T24gOC8yNC8wNywgRW1yZSBBeWT9biA8ZW1yYXlkaW5AZ21haWwuY29tPiB3cm90ZToKPiBoaS4g
aSdtIGEgbmV3IHVzZXIgb2YgbWF0cGxvdGxpYi4gaSd2ZSBzZWFyY2hlZCB0aHJvdWdoIHRoZSBh
cmhpZXZlcwo+IG9mIHRoZSBtYWlsaW5nIGxpc3QgYnV0IGNvdWxkbid0IGZpbmQgYSBxdWljayBz
b2x1dGlvbi4gaSBzaW1wbHkgbmVlZAo+IHRvIHJldmVyc2UgdGhlIHkgYXhpcyBvZiBteSBwbG90
cy4gZm9yIGV4YW1wbGUgaW4gYSByYW5nZSBvZiAwLTEwLCAwCj4gbXVzdCBzZWVtIGF0IHRoZSB0
b3Agb2YgdGhlIHBsb3Qgd2hlcmUgMTAgaXMgbG93ZXN0LCBuZWFyIHRvIHRoZSB4Cj4gYXhpcy4g
aXMgdGhlcmUgYSBxdWljayBzb2x1dGlvbiBmb3IgdGhpcy4gKHggYXhpcyBtdXN0IHN0YXkgYXMg
aXQgaXMpCgp5bGltKDEwLCAwKQoKSkRICg==

Alen Ribic wrote:
> Thanks Fred.
> 
> Thant did the trick. However now, when I have many plots on x axis,
> the last few plot shoot of the end of the x axis. It seems to start
> the plotting the middle move to the right. Do I just have to adjust
> the xlim on the axes[0]? I fiddled with the "align" parameter, set it
> to "center", on the bar function and it didn't do much.
> 
> -Alen
Hi Alen,
The align parameter sets whether the left, or the center of the bar 
should be aligned with the x-value you give for that bar. So the right 
hand edge of the bar would be at x+width (or x + (width/2) for center), 
and you'd need to set x_lim to include this. If I've misunderstood your 
problem can you post the code that's causing trouble?
Cheers,
Fred
ps.
gca() also gets the current axes object, which is marginally less typing 
than what I said before, so you can use:
gca().set_xlim([0.0, 10.0])
draw()
to re-scale the x-axis on the last thing you plotted to 0.0-10.0.

hi. i'm a new user of matplotlib. i've searched through the arhieves
of the mailing list but couldn't find a quick solution. i simply need
to reverse the y axis of my plots. for example in a range of 0-10, 0
must seem at the top of the plot where 10 is lowest, near to the x
axis. is there a quick solution for this. (x axis must stay as it is)
thanx...
-- 
EMRE AYDIN
emr...@gm...

Thanks Fred.
Thant did the trick. However now, when I have many plots on x axis,
the last few plot shoot of the end of the x axis. It seems to start
the plotting the middle move to the right. Do I just have to adjust
the xlim on the axes[0]? I fiddled with the "align" parameter, set it
to "center", on the bar function and it didn't do much.
-Alen
>
> On 8/24/07, Fred Ludlow <rf...@ca...> wrote:
> > Alen Ribic wrote:
> > > How do I set my vertical bar to be fixed width?
> >
> > By default, bars are created with a fixed width of 0.8, which can be
> > changed with the width keyword arg like so:
> >
> > bar(range(2), range(1,3), width = 0.5)
> >
> >
> > > Depending on amount of data on my x axis, the bars get created
> > > accordingly and the width gets adjusted to fit into the graph.
> > The width in pixels/proportion of axis is dependent on the range of the
> > x-axis, which you can set manually too.
> >
> > There's probably a more elegant way of doing this, but here's a quick
> > and dirty method for interactive mode (ipython/pylab):
> >
> > # Make a figure, catch the reference as f
> > f = figure(1)
> >
> > # Draw some bars
> > bar_list = bar(array([0,0.5,1.0]), array([1.0,2.0,3.0]), width = 0.25)
> >
> > # Get the subplot (f.axes is a list containing the subplots)
> > s = f.axes[0]
> >
> > # Set the x-axis limits of the subplot
> > s.set_xlim((-5.0, 5.0))
> >
> > # Redraw
> > draw()
> >
> >
> > Fred
> >
> > -------------------------------------------------------------------------
> > This SF.net email is sponsored by: Splunk Inc.
> > Still grepping through log files to find problems? Stop.
> > Now Search log events and configuration files using AJAX and a browser.
> > Download your FREE copy of Splunk now >> http://get.splunk.com/
> > _______________________________________________
> > Matplotlib-users mailing list
> > Mat...@li...
> > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> >
>

OK I see... nothing straightforward...
Best way for me is maybe to implement such a system myself:
The system would collect the information to be saved by kind of
introspection of the figure.
I'm planning to save data and plot properties in an hdf5 file. Kind of
inverted process will be used to restore the figure.
Of course I will not cover the whole possible figure case but only
what I'm currently using for my work.
If somebody did similar work and is eager to share or
if somebody have any suggestion please let me know.
Thanks for the great matplotlib and to its community !
Regards,
David

Alen Ribic wrote:
> How do I set my vertical bar to be fixed width?
By default, bars are created with a fixed width of 0.8, which can be
changed with the width keyword arg like so:
bar(range(2), range(1,3), width = 0.5)
> Depending on amount of data on my x axis, the bars get created
> accordingly and the width gets adjusted to fit into the graph.
The width in pixels/proportion of axis is dependent on the range of the
x-axis, which you can set manually too.
There's probably a more elegant way of doing this, but here's a quick 
and dirty method for interactive mode (ipython/pylab):
# Make a figure, catch the reference as f
f = figure(1)
# Draw some bars
bar_list = bar(array([0,0.5,1.0]), array([1.0,2.0,3.0]), width = 0.25)
# Get the subplot (f.axes is a list containing the subplots)
s = f.axes[0]
# Set the x-axis limits of the subplot
s.set_xlim((-5.0, 5.0))
# Redraw
draw()
Fred

Hi,
I'd like to open a high-quality image (600dpi) in matplotlib, add some
plots and save it as a postscript file. 
It seems that whatever I do, the input image gets scaled down
:-(
The same question was asked year ago. Has any progress been made since
then?
http://sourceforge.net/mailarchive/message.php?msg_id=200608311227.30594.dd55%40cornell.edu
Looking in the code of matplotlib, there is a constant of 72dpi
hardwired everywhere. For instance, backend_ps.py there is the line
	self.figure.dpi.set(72) # ignore the dpi kwarg
Would rising of this value to 600 help?
No offence to developers, I did not even try to understand the code, but
using the number 72 instead of using a variable seems to be a bad
programming practice to me. 
Petr Danecek

Le 2007年8月22日 18:21:40 -0700, Tom Vaughan a écrit:
> Why on the YellowDog 3 system would the x-axis show up as 0 - 2.5, and
> on the Ubuntu Feisty system would the x-axis show up as 2.2 - 2.4? I am
> attempting to resolve an autoscale problem elsewhere, and I must of
> screwed something up when I built matplotlib. But what?
Are you sure you have the same pref defined in conf files like 
~/.matplotlib/.matplotlibrc for example ?
-- 
Fabrice Silva
si...@cr...
06.15.59.07.61

On Tue, Aug 21, 2007 at 01:14:44PM +0100, Alex Pounds wrote:
> 4. I need to stick a footnote in the bottom corner of my charts. Can I do
> this natively in matplotlib or will I have to do it with a separate library
> afterwards?
For the sake of others searching the archives, here are the answers I've
found out for myself so far: 
4. You can do this with matplotlib: 
figtext(0.98, 0.05, u'©2007 Alexpounds.com', ha='right')
This puts a copyright notice in the bottom right of a chart. 
> 5. Can I have my xtick labels oriented vertically? 
Yep. Just pass rotation="vertical" in to xticks(). Easy. Though this may
overlap with the figtext above - I haven't got that bit sorted yet...
-- 
Alex Pounds (Creature) .~. http://www.alexpounds.com/
 /V\ http://www.ethicsgirls.com/
 // \\
"Variables won't; Constants aren't" /( )\
 ^`~'^

How do I set my vertical bar to be fixed width? Depending on amount of
data on my x axis, the bars get created accordingly and the width gets
adjusted to fit into the graph. If I have only 2 plots on the x axis
then the 2 bars get stretched across the entire graph. Looks very
ugly.
-Alen

> > 
> > I am but a humble newbie, but why not simply take your figure
> > object/reference and Pickle it (see
> > http://docs.python.org/lib/module-pickle.html)?
> > 
> 
> Won't work. Pickling only works for objects that have been designed for 
> it. Such design is not trivial for extension code, and has not been 
> done for mpl.
I keep coming back to this from time-to-time and have a go a poking
around in the mpl codebase but I've never figured out exactly what state
the matplotlib "Lazy Values" and "BinOps" store. 
If this internal state could be retrieved and then later restored, then
python can pickle the extension objects using the copy-reg module. It
might be enough if methods were added to the BinOp objects to retrieve
internal objects to which they link. You could then recursively walk
through the whole lazy-value tree. To make this work, you would need to
retrieve the underlying C++ objects with it's python wrapper object
intact (i.e. with the same IDs) so that python references to that object
created elsewhere still work.
While the concept of Lazy Values is easy enough to understand,
understanding how and where these objects fit in to the MPL architecture
is hard. I think a little documentation or explanation of this could go
a long way to helping implement this. If I could understand all the
places the Lazy Values are used, I'm keen to attempt the implementation
of pickle-ability. One problem for me is my C++ knowledge is ~ 0.
BC
> 
> Eric
> 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems? Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >> http://get.splunk.com/

Hi Eric,
On 23/08/07, Eric Emsellem <ems...@ob...> wrote:
> thanks a lot for this feedback!
>
> Your example is quite nice indeed. However there is something I may not
> have fully understand.
> If I use the example you sketch, I of course need to call the displayer
> class (right?), by doing something like:
>
> test = displayer()
>
> However, then I hit the same problem again: I would need to define ALL
> the commands I wish to go through WITHIN the displayer class (and more
> precisely into the "show_next" function), because any command put after
> the "test = displayer()" would be executed anyway without waiting that I
> finally hit the right mouse button.
I'm not sure why this is a problem. Yes, you need to put all your code
into the callback routines of displayer, but just think of it as
wrapping your program code inside the class. Your __main__ code then
simply becomes test = displayer()... and away you go.
Also, I realised shortly after sending my email that you can of course
do away with the offset class in this example, and put those methods
into the displayer class too, which will make things a little simpler.
> Or is there something I didn't catch which would allow me to go around this?
>
> (what my program is supposed to do at the end is to go through a series
> of (3xdataframes), and for each dataframe in turn, use the "offset"
> trick, then reinitialise everything and start with the next series...
> This seems to imply that I need to specify the full set of commands
> within the displayer class show_next function)
>
> thanks again
>
> Eric
Angus.
-- 
AJC McMorland, PhD Student
Physiology, University of Auckland

Alex Pounds wrote:
> On Thu, August 23, 2007 5:33 pm, David Tremouilles wrote:
>> I would like to save a matplotlib figure (data, title and axes label,
>> plot properties, ...) to load it later on for modification. Something
>> like figure.savelall("file.matplot") and later on do a
>> figure.loadall("file.matplot") using an empty figure.
> 
> I am but a humble newbie, but why not simply take your figure
> object/reference and Pickle it (see
> http://docs.python.org/lib/module-pickle.html)?
> 
Won't work. Pickling only works for objects that have been designed for 
it. Such design is not trivial for extension code, and has not been 
done for mpl.
Eric

On 8/23/07, Alex Pounds <al...@al...> wrote:
> On Thu, August 23, 2007 5:33 pm, David Tremouilles wrote:
> > I would like to save a matplotlib figure (data, title and axes label,
> > plot properties, ...) to load it later on for modification. Something
> > like figure.savelall("file.matplot") and later on do a
> > figure.loadall("file.matplot") using an empty figure.
> I am but a humble newbie, but why not simply take your figure
> object/reference and Pickle it (see
> http://docs.python.org/lib/module-pickle.html)?
Reasonable request, reasonable solution, but alas neither will work.
The mpl extension code doesn't support pickling.
JDH

On Thu, August 23, 2007 5:33 pm, David Tremouilles wrote:
> I would like to save a matplotlib figure (data, title and axes label,
> plot properties, ...) to load it later on for modification. Something
> like figure.savelall("file.matplot") and later on do a
> figure.loadall("file.matplot") using an empty figure.
I am but a humble newbie, but why not simply take your figure
object/reference and Pickle it (see
http://docs.python.org/lib/module-pickle.html)?
-- 
Alex Pounds (Creature) .~. http://www.alexpounds.com/
 /V\ http://www.ethicsgirls.com/
 // \\
"Variables won't; Constants aren't" /( )\
 ^`~'^

Hello,
 I would like to save a matplotlib figure (data, title and axes label,
plot properties, ...) to load it later on for modification. Something
like figure.savelall("file.matplot") and later on do a
figure.loadall("file.matplot") using an empty figure.
Did somebody already implement such a functionality?
If yes Is it available somewhere or are you eager to share it?
Thanks,
David

Hi
thanks a lot for this feedback!
Your example is quite nice indeed. However there is something I may not
have fully understand.
If I use the example you sketch, I of course need to call the displayer
class (right?), by doing something like:
test = displayer()
However, then I hit the same problem again: I would need to define ALL
the commands I wish to go through WITHIN the displayer class (and more
precisely into the "show_next" function), because any command put after
the "test = displayer()" would be executed anyway without waiting that I
finally hit the right mouse button.
Or is there something I didn't catch which would allow me to go around this?
(what my program is supposed to do at the end is to go through a series
of (3xdataframes), and for each dataframe in turn, use the "offset"
trick, then reinitialise everything and start with the next series...
This seems to imply that I need to specify the full set of commands
within the displayer class show_next function)
thanks again
Eric