Hi alexander,
I tryed yesterday to play a little with your script.
I suspect that in fact each time you draw a figure, it is added to a pool of
figure to be rendered.
Thus it goes slower and slower.
What I do in my scritpt is either update the datas or if the drawing is a
completely new one (like a polar ionstaed of log previously) I delete the
figure.
To update data use:
1/self._plot, = axe.plot(x, y, 'b', animated=self._animated)
Later:
Loop:
2/self._plot.set_ydata(self._value)
If you need to delete your figure (that is not your case, but who knows for
the future):
axe = fig.add_axes(self.getPosition())
self._fig.delaxes(self._axe)
To go even faster(x10) you have the blitting method, but set_ydata should be
sufficent.
Laurent
De : Alexander Hupfer [mailto:son...@gmail.com]
Envoyé : jeudi 7 janvier 2010 03:36
Cc : matplotlib-users@lists.sourceforge.net
Objet : Re: [Matplotlib-users] Colorbar embedding in qt4
I isolated the problem a bit more. For some reason drawing gets slower and
slower with each plot drawn.
from os import sys
from time import time
from PyQt4 import QtGui, QtCore
import matplotlib
matplotlib.use('QT4Agg')
from matplotlib.backends.backend_qt4agg import FigureCanvasQTAgg as
FigureCanvas
from matplotlib.figure import Figure
class MyMplCanvas(FigureCanvas):
"""Ultimately, this is a QWidget (as well as a FigureCanvasAgg,
etc.)."""
def __init__(self, parent=None, width=5, height=4, dpi=100):
fig = Figure(figsize=(width, height), dpi=dpi)
self.axes = fig.add_subplot(111)
# We want the axes cleared every time plot() is called
self.axes.hold(False)
self.compute_initial_figure()
#
FigureCanvas.__init__(self, fig)
self.setParent(parent)
FigureCanvas.setSizePolicy(self,
QtGui.QSizePolicy.Expanding,
QtGui.QSizePolicy.Expanding)
FigureCanvas.updateGeometry(self)
def compute_initial_figure(self):
pass
class MyDynamicMplCanvas(MyMplCanvas):
"""A canvas that updates itself every second with a new plot."""
def __init__(self, *args, **kwargs):
MyMplCanvas.__init__(self, *args, **kwargs)
timer = QtCore.QTimer(self)
QtCore.QObject.connect(timer, QtCore.SIGNAL("timeout()"),
self.update_figure)
timer.start(0)
self.firstrun = True
self.colorbar = None
self.time = time()
self.p = None
def compute_initial_figure(self):
pass
def update_figure(self):
self.p = self.axes.scatter([1,2,3],[4,5,6], c = range(3),
animated=True)
self.axes.set_xlabel('psi')
self.axes.set_ylabel('delta')
if self.firstrun == True:
self.colorbar = self.axes.figure.colorbar(self.p)
self.firstrun = False
self.colorbar.set_clim(vmin=0,vmax=2)
self.colorbar.draw_all()
self.colorbar.set_label('time [abtr. ]')
self.draw()
newtime = time()
print newtime - self.time
self.time = newtime
class ApplicationWindow(QtGui.QMainWindow):
def __init__(self):
QtGui.QMainWindow.__init__(self)
self.setAttribute(QtCore.Qt.WA_DeleteOnClose)
self.main_widget = QtGui.QWidget(self)
l = QtGui.QVBoxLayout(self.main_widget)
dc = MyDynamicMplCanvas(self.main_widget, width=5, height=4,
dpi=100)
l.addWidget(dc)
self.main_widget.setFocus()
self.setCentralWidget(self.main_widget)
qApp = QtGui.QApplication(sys.argv)
aw = ApplicationWindow()
aw.setWindowTitle("%s" % "Slow!")
aw.setGeometry(100,100,800,600)
aw.show()
sys.exit(qApp.exec_())
On Mon, Jan 4, 2010 at 3:36 PM, Alexander Hupfer <son...@gmail.com> wrote:
Ok, here is the code as a whole. I think it's still short enough to
ilustrate the problem. Just start it with the datafile "elips" as an
argument
http://dl.dropbox.com/u/226980/elipsometry.tar.gz
The timer shows how long each render cycle takes. The time seems to grow
with number of cycles rendered even when no more points are added (the
points are calculated with a continous rate in a background thread and are
about 300 total).
On Sun, Jan 3, 2010 at 8:16 PM, John Hunter <jdh2...@gmail.com> wrote:
On Sun, Jan 3, 2010 at 7:02 PM, Alexander Hupfer <son...@gmail.com> wrote:
> Thanks I got it fixed.
> This leads to the follow up question:
> What is the right way to keep an application responsive while the graph is
> drawn?
> Drawing a scatter plot with 300 points seems to take a while. I guess I
need
> to launch the drawing in another thread but don't know exactly how to do
> this and only find examples of doing calculations in the background and
not
> actual widget interaction.
You posted some real code and a traceback, which helped move the ball
forward. What we really need to see to help you though is a complete,
free-standing example, that we can run on our machines, which exposes
the problem.
JDH
------------------------------------------------------------------------------
This SF.Net email is sponsored by the Verizon Developer Community
Take advantage of Verizon's best-in-class app development support
A streamlined, 14 day to market process makes app distribution fast and easy
Join now and get one step closer to millions of Verizon customers
http://p.sf.net/sfu/verizon-dev2dev
_______________________________________________
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users
