[matplotlib-devel] new colormap
Hi All, As part of nipy[1] we have a spectral colormap which we use and would like to include it upstream as part of matplotlib. I'm working on a patch but before I submit it, I need some advice on how boilerplate.py should be used. I've made the required additions to _cm.py, pylab.py and boilerplate.py and when I run boilerplate.py it generates a spectral() function as it does for all the other colormaps. My question is, how should the output of boilerplate.py be included into pylab.py? Should I just cut and paste the output, or is there some automagic tool for doing it? Cheers, Tim Leslie [1] http://neuroimaging.scipy.org/ - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV ___ Matplotlib-devel mailing list Matplotlib-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
[matplotlib-devel] segfault in dynamic_demo.py
When running examples/dynamic_demo I get a segfault with the following backtrace. Does anyone have any thoughts on what might be causing this? I'm using python 2.4 .4c1 and the latest svn version of mpl/numpy/scipy. Cheers, Tim #0 0x004be54f in PyFrame_New () #1 0x00476004 in PyEval_EvalCodeEx () #2 0x004bf233 in PyClassMethod_New () #3 0x00413bf0 in PyObject_Call () #4 0x0046faf1 in PyEval_CallObjectWithKeywords () #5 0x2b041ffb412f in init_gobject () from /var/lib/python-support/python2.4/gtk-2.0/gobject/_gobject.so #6 0x2b042035218b in g_source_get_current_time () from /usr/lib/libglib-2.0.so.0 #7 0x2b0420351c84 in g_main_context_dispatch () from /usr/lib/libglib-2.0.so.0 #8 0x2b0420354acd in g_main_context_check () from /usr/lib/libglib-2.0.so.0 #9 0x2b0420354dda in g_main_loop_run () from /usr/lib/libglib-2.0.so.0 #10 0x2b0420a645f3 in gtk_main () from /usr/lib/libgtk-x11-2.0.so.0 #11 0x2b04295fb601 in wxEventLoop::Run () from /usr/lib/libwx_gtk2u_core-2.6.so.0 #12 0x2b042968411b in wxAppBase::MainLoop () from /usr/lib/libwx_gtk2u_core-2.6.so.0 #13 0x2b0428c88487 in wxPyApp::MainLoop () from /usr/lib/python2.4/site-packages/wx-2.6-gtk2-unicode/wx/_core_.so #14 0x2b0428ce928f in wxPyFileSystemHandler::FindFirst () from /usr/lib/python2.4/site-packages/wx-2.6-gtk2-unicode/wx/_core_.so #15 0x00413bf0 in PyObject_Call () #16 0x00473fd0 in PyEval_EvalFrame () #17 0x004767d6 in PyEval_EvalCodeEx () #18 0x004bf233 in PyClassMethod_New () #19 0x00413bf0 in PyObject_Call () #20 0x00419930 in PyClass_IsSubclass () #21 0x00413bf0 in PyObject_Call () #22 0x00472619 in PyEval_EvalFrame () #23 0x00475546 in PyEval_EvalFrame () #24 0x004767d6 in PyEval_EvalCodeEx () #25 0x00474a5a in PyEval_EvalFrame () #26 0x004767d6 in PyEval_EvalCodeEx () #27 0x00476882 in PyEval_EvalCode () #28 0x0049b1e2 in PyRun_FileExFlags () #29 0x0049b3e0 in PyRun_SimpleFileExFlags () #30 0x00410b9a in Py_Main () #31 0x2b041fd400c4 in __libc_start_main () from /lib/libc.so.6 #32 0x00410079 in _start () - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV ___ Matplotlib-devel mailing list Matplotlib-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
Re: [matplotlib-devel] new colormap
On 12/20/06, John Hunter <[EMAIL PROTECTED]> wrote:
>>>>> "Tim" == Tim Leslie <[EMAIL PROTECTED]> writes:
Tim> colormaps. My question is, how should the output of
Tim> boilerplate.py be included into pylab.py? Should I just cut
Tim> and paste the output, or is there some automagic tool for
Tim> doing it?
Yep, just cut the stuff below
### Do not edit below this point
and paste in the boilerplate output.
Thanks, that all seems to have worked fine. I've attached the patch
with this email.
John, I was wondering if I have svn write access. I seem to recall I
had it a year or two ago, but I can't find any evidence to back this
up, so maybe I'm mistaken. Could you check for me?
Cheers,
Tim
JDH
Index: lib/matplotlib/_cm.py
===
--- lib/matplotlib/_cm.py (revision 2941)
+++ lib/matplotlib/_cm.py (working copy)
@@ -14,6 +14,7 @@
+
_bone_data = {'red': ((0., 0., 0.),(1.0, 1.0, 1.0)),
'green': ((0., 0., 0.),(1.0, 1.0, 1.0)),
'blue': ((0., 0., 0.),(1.0, 1.0, 1.0))}
@@ -338,6 +339,41 @@
'green': ((0., 0., 0.),(1.0, 1.0, 1.0)),
'blue': ((0., 1., 1.),(1.0, 0.5, 0.5))}
+_spectral_data = {'red': [(0.0, 0.0, 0.0), (0.05, 0.4667, 0.4667),
+ (0.10, 0.5333, 0.5333), (0.15, 0.0, 0.0),
+ (0.20, 0.0, 0.0), (0.25, 0.0, 0.0),
+ (0.30, 0.0, 0.0), (0.35, 0.0, 0.0),
+ (0.40, 0.0, 0.0), (0.45, 0.0, 0.0),
+ (0.50, 0.0, 0.0), (0.55, 0.0, 0.0),
+ (0.60, 0.0, 0.0), (0.65, 0.7333, 0.7333),
+ (0.70, 0.9333, 0.9333), (0.75, 1.0, 1.0),
+ (0.80, 1.0, 1.0), (0.85, 1.0, 1.0),
+ (0.90, 0.8667, 0.8667), (0.95, 0.80, 0.80),
+ (1.0, 0.80, 0.80)],
+ 'green': [(0.0, 0.0, 0.0), (0.05, 0.0, 0.0),
+(0.10, 0.0, 0.0), (0.15, 0.0, 0.0),
+(0.20, 0.0, 0.0), (0.25, 0.4667, 0.4667),
+(0.30, 0.6000, 0.6000), (0.35, 0.6667, 0.6667),
+(0.40, 0.6667, 0.6667), (0.45, 0.6000, 0.6000),
+(0.50, 0.7333, 0.7333), (0.55, 0.8667, 0.8667),
+(0.60, 1.0, 1.0), (0.65, 1.0, 1.0),
+(0.70, 0.9333, 0.9333), (0.75, 0.8000, 0.8000),
+(0.80, 0.6000, 0.6000), (0.85, 0.0, 0.0),
+(0.90, 0.0, 0.0), (0.95, 0.0, 0.0),
+(1.0, 0.80, 0.80)],
+ 'blue': [(0.0, 0.0, 0.0), (0.05, 0.5333, 0.5333),
+ (0.10, 0.6000, 0.6000), (0.15, 0.6667, 0.6667),
+ (0.20, 0.8667, 0.8667), (0.25, 0.8667, 0.8667),
+ (0.30, 0.8667, 0.8667), (0.35, 0.6667, 0.6667),
+ (0.40, 0.5333, 0.5333), (0.45, 0.0, 0.0),
+ (0.5, 0.0, 0.0), (0.55, 0.0, 0.0),
+ (0.60, 0.0, 0.0), (0.65, 0.0, 0.0),
+ (0.70, 0.0, 0.0), (0.75, 0.0, 0.0),
+ (0.80, 0.0, 0.0), (0.85, 0.0, 0.0),
+ (0.90, 0.0, 0.0), (0.95, 0.0, 0.0),
+ (1.0, 0.80, 0.80)]}
+
+
autumn = colors.LinearSegmentedColormap('autumn', _autumn_data, LUTSIZE)
bone = colors.LinearSegmentedColormap('bone ', _bone_data, LUTSIZE)
cool = colors.LinearSegmentedColormap('cool', _cool_data, LUTSIZE)
@@ -352,10 +388,10 @@
spring = colors.LinearSegmentedColormap('spring', _spring_data, LUTSIZE)
summer = colors.LinearSegmentedColormap('summer', _summer_data, LUTSIZE)
winter = colors.LinearSegmentedColormap('winter', _winter_data, LUTSIZE)
+spectral = colors.LinearSegmentedColormap('spectral', _spectral_data, LUTSIZE)
-
datad = {
'autumn': _autumn_data,
'bone': _bone_data,
@@ -370,7 +406,8 @@
'prism': _prism_data,
'spring': _spring_data,
'summer': _summer_data,
-'winter': _winter_data
+'winter': _winter_data,
+'spectral': _spectral_data
}
# 34 colormaps based on color specifications and designs
Index: lib/matplotlib/pylab.py
===
--- lib/matplotlib/pylab.py (revision 2941)
+++ lib/matplotlib/pylab.py (working copy)
@@ -102,6 +102,7 @@
spring - set the default colormap to spring
summer - set the default colormap to summer
winter - set the default colormap to winter
+ spectral - set the default color
Re: [matplotlib-devel] [Fwd: Re: [Distutils] Confusion about the effect of eggs on import]
On 2/11/07, Gael Varoquaux <[EMAIL PROTECTED]> wrote: > On Sat, Feb 10, 2007 at 08:04:38AM -0600, John Hunter wrote: > > I am inclined to consider ripping out the __init__ stuff into a > > "config" module or something like that. > > If my understanding is right, enthought does this with an "api" module. > It might be worth going in the same direction. > And for what it's worth this is also the direction nipy is taking, having an api.py in each package. Tim > My 2 cents, > > Gaël > > -- > Gael Varoquaux, > Groupe d'optique atomique, > Laboratoire Charles Fabry de l'Institut d'Optique > Campus Polytechnique, RD 128 > 91127 Palaiseau cedex FRANCE > NEW Phone number > Tel : 33 (0) 1 64 54 33 49 - Fax : 33 (0) 1 64 54 31 01 > Labs: 33 (0) 1 64 54 33 63 - 33 (0) 1 64 54 33 62 > > - > Using Tomcat but need to do more? Need to support web services, security? > Get stuff done quickly with pre-integrated technology to make your job easier. > Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo > http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 > ___ > Matplotlib-devel mailing list > Matplotlib-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/matplotlib-devel > - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier. Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 ___ Matplotlib-devel mailing list Matplotlib-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
Re: [matplotlib-devel] [Fwd: Re: [Distutils] Confusion about the effect of eggs on import]
On 2/11/07, John Hunter <[EMAIL PROTECTED]> wrote: > On 2/10/07, Tim Leslie <[EMAIL PROTECTED]> wrote: > > > And for what it's worth this is also the direction nipy is taking, > > having an api.py in each package. > > I don't have a problem with this if it makes sense to other people and > helps with consistency across packages, but I don't find the mnemonic > particularly clear. Eg > > from matplotlib.config import rc, verbose > > makes more sense to me than > > from matplotlib.api import rc, verbose > > since I consider these (and most other stuff in __init__) to be part > of the configuration, and a lot of stuff that doesn't live there > (Figure, Axes, Line2D) to be part of the API. What is the > enthought/nipy rationale for calling it api? > Sorry, I think I might have given the wrong impression. Our use of an api.py is for having a single location to import classes equivalent to Figure, Axes, etc from. Your suggestion of having a matplotlib.config for things like rc and verbose makes a lot of sense I think. Tim > JDH > - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier. Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 ___ Matplotlib-devel mailing list Matplotlib-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
[matplotlib-devel] Two small patches
Hi All,
I've attached a patch which addresses two small issues. The first is a
bug in patches.py. When doing 3d scatter plots
xs, ys = zip(*self.xy) (line 480)
crashes, as the right hand side expands to 3 values.
The other is a speedup, normalising an entire array at once, rather
than element by element, speeding up the attached script by a factor
of 6x (and upgrading my actual application from 'too slow' to 'quite
reasonable' :-)
Cheers,
Tim
Index: lib/matplotlib/art3d.py
===
--- lib/matplotlib/art3d.py (revision 3193)
+++ lib/matplotlib/art3d.py (working copy)
@@ -369,7 +369,7 @@
"""Modify the alphas of the color list according to depth"""
colors = get_colors(colors,len(zs))
norm = Normalize(min(zs),max(zs))
-sats = nx.array([1-norm(z)*0.7 for z in zs])
+sats = 1 - norm(zs)*0.7
colors = [(c[0],c[1],c[2],c[3]*s) for c,s in zip(colors,sats)]
return colors
Index: lib/matplotlib/patches.py
===
--- lib/matplotlib/patches.py (revision 3193)
+++ lib/matplotlib/patches.py (working copy)
@@ -480,7 +480,7 @@
def get_verts(self):
-xs, ys = zip(*self.xy)
+xs, ys = zip(*self.xy)[:2]
xs = self.convert_xunits(xs)
ys = self.convert_yunits(ys)
return zip(xs, ys)
import random
import pylab as p
import matplotlib.axes3d as p3
def main():
fig = p.figure()
ax = p3.Axes3D(fig)
pts = 7500
X = range(pts)
Y = range(pts)
Z = range(pts)
ax.scatter3D(X, Y, Z)
p.savefig("tmp.png")
if __name__ == '__main__':
import hotshot, hotshot.stats
prof = hotshot.Profile("stones.prof")
benchtime = prof.runcall(main)
prof.close()
stats = hotshot.stats.load("stones.prof")
stats.strip_dirs()
stats.sort_stats('cumulative', 'calls')
stats.print_stats(20)
-
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV___
Matplotlib-devel mailing list
Matplotlib-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
Re: [matplotlib-devel] Two small patches
On 4/10/07, John Hunter <[EMAIL PROTECTED]> wrote: > On 4/9/07, Tim Leslie <[EMAIL PROTECTED]> wrote: > > > > > The other is a speedup, normalising an entire array at once, rather > > than element by element, speeding up the attached script by a factor > > of 6x (and upgrading my actual application from 'too slow' to 'quite > > reasonable' :-) > > Thanks Tim, I just committed this. I see that my sinister plan of > releasing the 3d stuff w/o support in hopes of reeling in contributers > is starting to work Well I've got a bit of work to do which involves 3D stuff in the coming week or so, so if I run up against anything else I'll keep the patches coming. Thanks for the quick commit. Cheers, Tim > > > - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV ___ Matplotlib-devel mailing list Matplotlib-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
Re: [matplotlib-devel] Two small patches
On 4/10/07, John Hunter <[EMAIL PROTECTED]> wrote: On 4/9/07, Tim Leslie <[EMAIL PROTECTED]> wrote: > > The other is a speedup, normalising an entire array at once, rather > than element by element, speeding up the attached script by a factor > of 6x (and upgrading my actual application from 'too slow' to 'quite > reasonable' :-) Thanks Tim, I just committed this. I see that my sinister plan of releasing the 3d stuff w/o support in hopes of reeling in contributers is starting to work Here's another patch. This one cleans up the line2d_seg_dist function and also speeds it up by a factor of ~2. Cheers, Tim Index: proj3d.py === --- proj3d.py (revision 3198) +++ proj3d.py (working copy) @@ -51,52 +51,28 @@ x0,y0 = p return abs((a*x0 + b*y0 + c)/nx.sqrt(a**2+b**2)) -def dist2d(p0,p1): -"""distance between two points""" -try: -p = p0-p1 -return nx.sqrt(sum(p**2)) -except ValueError: -print p0,p1 -raise -def line2d_seg_dist(p0,p1, p): -"""distance(s) from line defined by p0 - p1 to point(s) p +def line2d_seg_dist(p1,p2, p0): +"""distance(s) from line defined by p1 - p2 to point(s) p0 -p[0] = x(s) -p[1] = y(s) +p0[0] = x(s) +p0[1] = y(s) -intersection point p = p0 + u*(p1-p0) +intersection point p = p1 + u*(p2-p1) and intersection point lies within segement if u is between 0 and 1 """ - -p,p0,p1 = map(nx.asarray,(p[:2],p0[:2],p1[:2])) -x,y = p -x0,y0 = p0 -x1,y1 = p1 -# -u = ((x-x0)*(x1-x0)+(y-y0)*(y1-y0))/(dist2d(p0,p1)*dist2d(p0,p1)) -# -def dist(p,p0,p1,u): -if u > 0 and u < 1: -pi = p0 + u*(p1-p0) -return dist2d(p,pi) -else: -return nx.minimum(dist2d(p,p0),dist2d(p,p1)) - -if not iterable(u): -return dist(p,p0,p1,u) -else: -# for each point calculate dist -pt = nx.transpose(p) -return nx.array([dist(pe,p0,p1,ue) for pe,ue in zip(pt,u)]) -def iterable(v): -try: -len(v) -return True -except TypeError: -return False +x21 = p2[0] - p1[0] +y21 = p2[1] - p1[1] +x01 = nx.asarray(p0[0]) - p1[0] +y01 = nx.asarray(p0[1]) - p1[1] + +u = (x01*x21 + y01*y21)/float(abs(x21**2 + y21**2)) +u = nx.clip(u, 0, 1) +d = nx.sqrt((x01 - u*x21)**2 + (y01 - u*y21)**2) + +return d + def test_lines_dists(): ax = pylab.gca() - Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV___ Matplotlib-devel mailing list Matplotlib-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
[matplotlib-devel] [patch] RegularPolyCollection.get_transformed_patches speed-up
Hi All, I've attached a patch which optimizes a particular case of the RegularPolyCollection.get_transformed_patches method. This is particularly useful when using a picker on a scatter plot with ~40,000 points (as I happen to be doing) and cuts the time spent in this function from ~4s to ~1s, which accounts for about 50% of the lag in my particular use case. Similar improvements might be possible in the N > 1 case, but I don't have a) a data set or b) the time to work them out with, so hopefully someone is happy to check in this patch as it stands, as I think the N==1 is probably the common case. If there are any problems with the patch let me know and I'll try to fix it up. It's against latest svn (3262) Cheers, Tim Index: lib/matplotlib/collections.py === --- lib/matplotlib/collections.py (revision 3262) +++ lib/matplotlib/collections.py (working copy) @@ -153,7 +153,7 @@ if not self.pickable(): return ind = [] x, y = mouseevent.x, mouseevent.y -for i, thispoly in enumerate(self.get_transformed_patches()): +for i, thispoly in enumerate(self.get_transformed_patches()): inside = nxutils.pnpoly(x, y, thispoly) if inside: ind.append(i) if len(ind): @@ -167,7 +167,6 @@ The ith element in the returned sequence is a list of x,y vertices defining the ith polygon """ - verts = self._verts offsets = self._offsets usingOffsets = offsets is not None @@ -451,13 +450,19 @@ __init__.__doc__ = dedent(__init__.__doc__) % kwdocd def get_transformed_patches(self): - + xverts, yverts = zip(*self._verts) xverts = asarray(xverts) yverts = asarray(yverts) sizes = sqrt(asarray(self._sizes)*self._dpi.get()/72.0) Nsizes = len(sizes) transOffset = self.get_transoffset() + +# optimize this case for an approx 4x speedup +if Nsizes == 1: +xy = asarray([xverts, yverts]).T * sizes[0] +polys = [xy + offset for offset in transOffset.seq_xy_tups(self._offsets)] + polys = [] for i, loc in enumerate(self._offsets): xo,yo = transOffset.xy_tup(loc) - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/___ Matplotlib-devel mailing list Matplotlib-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
Re: [matplotlib-devel] [patch] RegularPolyCollection.get_transformed_patches speed-up
On 5/10/07, Eric Firing <[EMAIL PROTECTED]> wrote: > Tim, > > Based on a *very superficial* quick look, I have two comments: > > 1) Since the plan is (or was?) to have one more release before > specializing to numpy-only support, the ".T" won't work at present. > > 2) In the following (possibly mangled by mailer), > > > +if Nsizes == 1: > > +xy = asarray([xverts, yverts]).T * sizes[0] > > +polys = [xy + offset for offset in > transOffset.seq_xy_tups(self._offsets)] > > I think you can gain additional speedup by eliminating the list > comprehension. Instead of using seq_xy_tups you can use numerix_xy to > return a 2-D array (with columns x and y), and simply add that to xy. > Thanks for the feedback Eric, I'll rework the patch with your comments in mind. I also just noticed a rather glaring bug, in that I don't actually have a 'return polys' after I calculate them in the special case, which is somewhat embarrassing. I'll fix all this up and submit a new patch later today. Cheers, Tim > Eric > > > Tim Leslie wrote: > > Hi All, > > > > I've attached a patch which optimizes a particular case of the > > RegularPolyCollection.get_transformed_patches method. This is > > particularly useful when using a picker on a scatter plot with ~40,000 > > points (as I happen to be doing) and cuts the time spent in this > > function from ~4s to ~1s, which accounts for about 50% of the lag in > > my particular use case. > > > > Similar improvements might be possible in the N > 1 case, but I don't > > have a) a data set or b) the time to work them out with, so hopefully > > someone is happy to check in this patch as it stands, as I think the > > N==1 is probably the common case. > > > > If there are any problems with the patch let me know and I'll try to > > fix it up. It's against latest svn (3262) > > > > Cheers, > > > > Tim > > > > > > > > > > Index: lib/matplotlib/collections.py > > === > > --- lib/matplotlib/collections.py (revision 3262) > > +++ lib/matplotlib/collections.py (working copy) > > @@ -153,7 +153,7 @@ > > if not self.pickable(): return > > ind = [] > > x, y = mouseevent.x, mouseevent.y > > -for i, thispoly in enumerate(self.get_transformed_patches()): > > +for i, thispoly in enumerate(self.get_transformed_patches()): > > inside = nxutils.pnpoly(x, y, thispoly) > > if inside: ind.append(i) > > if len(ind): > > @@ -167,7 +167,6 @@ > > The ith element in the returned sequence is a list of x,y > > vertices defining the ith polygon > > """ > > - > > verts = self._verts > > offsets = self._offsets > > usingOffsets = offsets is not None > > @@ -451,13 +450,19 @@ > > __init__.__doc__ = dedent(__init__.__doc__) % kwdocd > > > > def get_transformed_patches(self): > > - > > + > > xverts, yverts = zip(*self._verts) > > xverts = asarray(xverts) > > yverts = asarray(yverts) > > sizes = sqrt(asarray(self._sizes)*self._dpi.get()/72.0) > > Nsizes = len(sizes) > > transOffset = self.get_transoffset() > > + > > +# optimize this case for an approx 4x speedup > > +if Nsizes == 1: > > +xy = asarray([xverts, yverts]).T * sizes[0] > > +polys = [xy + offset for offset in > > transOffset.seq_xy_tups(self._offsets)] > > + > > polys = [] > > for i, loc in enumerate(self._offsets): > > xo,yo = transOffset.xy_tup(loc) > > > > > > > > > > - > > This SF.net email is sponsored by DB2 Express > > Download DB2 Express C - the FREE version of DB2 express and take > > control of your XML. No limits. Just data. Click to get it now. > > http://sourceforge.net/powerbar/db2/ > > > > > > > > > > ___ > > Matplotlib-devel mailing list > > Matplotlib-devel@lists.sourceforge.net > > https://lists.sourceforge.net/lists/listinfo/matplotlib-devel > > - This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ ___ Matplotlib-devel mailing list Matplotlib-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
Re: [matplotlib-devel] [patch] RegularPolyCollection.get_transformed_patches speed-up
Hi Eric, I just checked out your change and it works another 0.2s faster than what I had for my example, so I'm more than happy with the final result. Thanks for taking the time to put my hazy ideas into practice the right way :-) Cheers, Tim On 5/10/07, Eric Firing <[EMAIL PROTECTED]> wrote: > Tim, > > I couldn't resist; here is a numerix version of the method, which I > committed to svn: > > def get_transformed_patches(self): > # Shouldn't need all these calls to asarray; > # the variables should be converted when stored. > # Similar speedups with numerix should be attainable > # in many other places. > verts = asarray(self._verts) > offsets = asarray(self._offsets) > Npoly = len(offsets) > scales = sqrt(asarray(self._sizes)*self._dpi.get()/72.0) > Nscales = len(scales) > if Nscales >1: > scales = resize(scales, (Npoly, 1, 1)) > transOffset = self.get_transoffset() > xyo = transOffset.numerix_xy(offsets) > polys = scales * verts + xyo[:, newaxis, :] > return polys > > It is faster even when there are only three polygons; and as the comment > indicates, it would be even faster with additional numpification of the > entire class. (Present version with 3 polys, 161 usecs vs 241 usecs; > with 1000, 6.4 msec vs 47.3 msec.) > > It is also only lightly tested. I hope I haven't overlooked some > situation that would work with the old version but not with this one. > > Eric > > Tim Leslie wrote: > > On 5/10/07, Eric Firing <[EMAIL PROTECTED]> wrote: > >> Tim, > >> > >> Based on a *very superficial* quick look, I have two comments: > >> > >> 1) Since the plan is (or was?) to have one more release before > >> specializing to numpy-only support, the ".T" won't work at present. > >> > >> 2) In the following (possibly mangled by mailer), > >> > >> > +if Nsizes == 1: > >> > +xy = asarray([xverts, yverts]).T * sizes[0] > >> > +polys = [xy + offset for offset in > >> transOffset.seq_xy_tups(self._offsets)] > >> > >> I think you can gain additional speedup by eliminating the list > >> comprehension. Instead of using seq_xy_tups you can use numerix_xy to > >> return a 2-D array (with columns x and y), and simply add that to xy. > >> > > > > Thanks for the feedback Eric, I'll rework the patch with your comments > > in mind. I also just noticed a rather glaring bug, in that I don't > > actually have a 'return polys' after I calculate them in the special > > case, which is somewhat embarrassing. I'll fix all this up and submit > > a new patch later today. > > > > Cheers, > > > > Tim > > > >> Eric > >> > >> > >> Tim Leslie wrote: > >> > Hi All, > >> > > >> > I've attached a patch which optimizes a particular case of the > >> > RegularPolyCollection.get_transformed_patches method. This is > >> > particularly useful when using a picker on a scatter plot with ~40,000 > >> > points (as I happen to be doing) and cuts the time spent in this > >> > function from ~4s to ~1s, which accounts for about 50% of the lag in > >> > my particular use case. > >> > > >> > Similar improvements might be possible in the N > 1 case, but I don't > >> > have a) a data set or b) the time to work them out with, so hopefully > >> > someone is happy to check in this patch as it stands, as I think the > >> > N==1 is probably the common case. > >> > > >> > If there are any problems with the patch let me know and I'll try to > >> > fix it up. It's against latest svn (3262) > >> > > >> > Cheers, > >> > > >> > Tim > >> > > >> > > >> > > >> > >> > > >> > Index: lib/matplotlib/collections.py > >> > === > >> > --- lib/matplotlib/collections.py (revision 3262) > >> > +++ lib/matplotlib/collections.py (working copy) > >> > @@ -153,7 +153,7 @@ > >> > if not self.pickable(): return > >> > ind = [] > >> > x, y = mouseevent.x, mouseevent.y > >> > -
