Hi Gökhan, Hi list, Hi experts on masked arrays, I run the attached script (copied from your last two emails) and I can use x_new, y_new after closing the figures (where x_new and y_new are the values selected in the last selection).
I don't know, what is going wrong with your script on your system.
### my output ##########################
In [1]: run rect.py
startposition : (0.276210, 0.592969)
endposition : (0.542339, 0.309375)
used button : 1
startposition : (0.745968, 0.747656)
endposition : (0.745968, 0.747656)
used button : 1
startposition : (0.645161, 0.761719)
endposition : (0.806452, 0.609375)
used button : 3
startposition : (0.808468, 0.607031)
endposition : (0.808468, 0.604688)
used button : 3
startposition : (0.086694, 0.166406)
endposition : (0.137097, 0.070313)
used button : 1
startposition : (0.921371, 0.850781)
endposition : (0.987903, 0.728906)
used button : 1
In [2]: x_new
Out[2]:
array([ 0.92929293, 0.93939394, 0.94949495, 0.95959596, 0.96969697,
0.97979798])
In [3]: y_new
Out[3]:
array([ 0.80119703, 0.8072005 , 0.81312162, 0.81895978, 0.82471437,
0.83038482])
In [4]:
#############
Please see my comments below.
On Monday 20 April 2009 23:16:37 Gökhan SEVER wrote:
> Thanks for elegant trick Matthias.
>
> I have modified onselect function following your suggestions, and it is
> working as I wanted it to be. Select a portion and get a zoomed view in a
> new figure.
>
> def onselect(eclick, erelease):
> # eclick and erelease are matplotlib events at press and release
> global x_new
> global y_new
> print ' startposition : (%f, %f)' % (eclick.xdata, eclick.ydata)
> print ' endposition : (%f, %f)' % (erelease.xdata, erelease.ydata)
> print ' used button : ', eclick.button
> xmin = min(eclick.xdata, erelease.xdata)
> xmax = max(eclick.xdata, erelease.xdata)
> ymin = min(eclick.ydata, erelease.ydata)
> ymax = max(eclick.ydata, erelease.ydata)
> indices = (x >= xmin) & (x <= xmax) & (y >= ymin) & (y <= ymax)
> x_new = x[indices]
> y_new = y[indices]
> fig_new = figure()
> plot(x_new, y_new)
> fig_new.show()
>
> There is still a minute point, that I would like mention again. Even though
> I used global style variables I can't still see them when I quit the
> program.
>
> What I do is run the file within IPython by run command. In my case the
> file is 'rect.py.' and I do run rect.py. When I quit and do whos querry, I
> can't see global x_new nor y_new at the resulting lines:
>
> In [13]: whos
> Variable Type Data/Info
> --------------------------------------------
> RectangleSelector classobj matplotlib.widgets.RectangleSelector
> ax AxesSubplot Axes(0.125,0.1;0.775x0.8)
> fig function <function figure at 0xa752c6c>
> onselect function <function onselect at 0xac0ec34>
> toggle_selector function <function toggle_selector at 0xac0ef0c>
> x ndarray 100: 100 elems, type `float64`, 800
> bytes y ndarray 100: 100 elems, type `float64`,
> 800 bytes
>
>
> For the masking, I was planning to use a masking scheme as given below. Let
> say I have an array which "a"
>
> In [14]: a = arange(5)
>
> In [15]: a
> Out[15]: array([0, 1, 2, 3, 4])
>
> and my secondary array is "b"
>
> In [16]: b = array([2,3])
>
> What I want to do is to mask a with b values and get an array of:
>
> [array([False, False, True, True, False], dtype=bool)]
>
> That is just an manually created array. I still don't know how to do this
> programmatically in Pythonic fashion.
As I mentioned before I'm not familiar with numpys masked arrays. In this list
I read several comments on them, but I never used them. But I think they are
what you are after. Maybe someone else can give you a hint or you try numpys
online documentation.
All I could provide is again a little manually work using a for-loop. So
something like
mask = zeros(len(a), dtype=bool)
for index in xrange(len(a)): # run through array a
if a[index] in b:
mask[index] = True
print mask # gives array([False, False, True, True, False], dtype=bool)
Furthermore I'm not a Python expert and therefore I'm not the guy to ask for
programing in "Pythonic fashion". Maybe some else can give a comment?
best regards Matthias
> Again thanks for your help.
>
> Gökhan
>
>
> On Mon, Apr 20, 2009 at 4:09 AM, Matthias Michler
>
> <matthiasmich...@gmx.net>wrote:
> > Hi Gökhan,
> >
> > On Friday 17 April 2009 20:21:00 Gökhan SEVER wrote:
> > > Thanks for the pointer Matthias,
> > >
> > > That is exactly what I have been looking for.
> > >
> > > I use the code from the RectangleSelector class help with your
> > > suggested code. I know that I have to update y-axis accordingly to x
> > > values such
> >
> > that
> >
> > > their positions and sizes must much so that I can plot them in a new
> >
> > plot.
> >
> > > And I know that the answer lies in a mask; I have to create a mask from
> > > x_new and apply it to y. Do you have any hint on this?
> >
> > Unfortunately I'm not familiar with numpy masks, but what I would do is:
> >
> > xmin = min(eclick.xdata, erelease.xdata)
> > xmax = max(eclick.xdata, erelease.xdata)
> > ymin = min(eclick.ydata, erelease.ydata)
> > ymax = max(eclick.ydata, erelease.ydata)
> > # indices inside x-range
> > indices = (x>= xmin) & (x <= xmax)
> > # OR: indices for data inside the selected rectangle
> > #indices = (x>= xmin) & (x <= xmax) & (y>=ymin) & (y<=ymax)
> > xnew = x[indices]
> > ynew = y[indices]
> >
> > > Another point is do you have any idea how to save values from inside
> > > onselect action?
> >
> > What do you mean by saving?
> > Saving to disk?
> > # for ascii format I use:
> > from scipy.io import write_array
> >
> > # for numpy arrays you can use
> > import numpy as np
> > a = np.arange(10)
> > a.tofile # Write array to a file as text or binary.
> >
> > and I think there is also some Matplotlib function for this. In the
> > module matplotlib.mlab, which also allows reading different types of
> > formatted data.
> >
> > If you think of saving inside the program. You need a global variable
> > (statement "global x" at the beginning of onselect) otherwise all
> > variables are deleted at the end of each onselect call.
> >
> > > For some reason my ipython session doesn't remember values after I run
> >
> > the
> >
> > > given script:
> >
> > see three lines above, please.
> >
> > best regards Matthias
> >
> > > from matplotlib.widgets import RectangleSelector
> > > from pylab import *
> > >
> > > def onselect(eclick, erelease):
> > > # eclick and erelease are matplotlib events at press and release
> > > print ' startposition : (%f, %f)' % (eclick.xdata, eclick.ydata)
> > > print ' endposition : (%f, %f)' % (erelease.xdata,
> > > erelease.ydata) print ' used button : ', eclick.button
> > > xmin = min(eclick.xdata, erelease.xdata)
> > > xmax = max(eclick.xdata, erelease.xdata)
> > > ymin = min(eclick.ydata, erelease.ydata)
> > > ymax = max(eclick.ydata, erelease.ydata)
> > > x_new = x[(x>= xmin) & (x <= xmax)]
> > > #mask = [x == x_new[i] for i in range(len(x_new))]
> > > #print mask
> > > #print len(x_new)
> > > #print len(y_new)
> > > #fig_new = figure()
> > > #plot(x_new, y_new)
> > > #fig_new.show()
> > >
> > > def toggle_selector(event):
> > > print ' Key pressed.'
> > > if event.key in ['Q', 'q'] and toggle_selector.RS.active:
> > > print ' RectangleSelector deactivated.'
> > > toggle_selector.RS.set_active(False)
> > > if event.key in ['A', 'a'] and not toggle_selector.RS.active:
> > > print ' RectangleSelector activated.'
> > > toggle_selector.RS.set_active(True)
> > >
> > > x = arange(100)/(99.0)
> > > y = sin(x)
> > > fig = figure
> > > ax = subplot(111)
> > > ax.plot(x,y)
> > >
> > > toggle_selector.RS = RectangleSelector(ax, onselect, drawtype='box')
> > > connect('key_press_event', toggle_selector)
> > > show()
> > >
> > > Gökhan
> > >
> > >
> > > On Fri, Apr 17, 2009 at 2:31 AM, Matthias Michler
> > >
> > > <matthiasmich...@gmx.net>wrote:
> > > > Hi Gökhan,
> > > >
> > > > I recommend you to use matplotlib.widgets.RectangleSelector instead
> > > > of the zoom functionality to select the data (An example can be found
> > > > at
> >
> > http://matplotlib.sourceforge.net/examples/widgets/rectangle_selector.htm
> >
> > > >l). This will return you the x and y-coordinate of button press and
> >
> > button
> >
> > > > release
> > > > event and with that you can take a portion of your data.
> > > > Something like the following could be a starting point:
> > > > x_min = min(eclick.xdata, erelease.xdata)
> > > > x_max = max(eclick.xdata, erelease.xdata)
> > > > x_new = x[(x>= x_min) & (x <= x_max)]
> > > >
> > > > where eclick and erelease correspond to the click and release event
> > > > of the rectangle selection (see the example below).
> > > >
> > > > Opening a new figure after show can be achieved by:
> > > >
> > > > fig_new = plt.figure()
> > > > # some plotting
> > > > fig_new.show() # show up the new figure
> > > >
> > > >
> > > > best regards Matthias
> > > >
> > > >
> > > > yet another example for the usage of the RectangleSelector copied
> > > > from its class documentation:
> > > >
> > > > """
> > > > Select a min/max range of the x axes for a matplotlib Axes
> > > >
> > > > Example usage::
> > > >
> > > > from matplotlib.widgets import RectangleSelector
> > > > from pylab import *
> > > >
> > > > def onselect(eclick, erelease):
> > > > 'eclick and erelease are matplotlib events at press and
> >
> > release'
> >
> > > > print ' startposition : (%f, %f)' % (eclick.xdata,
> >
> > eclick.ydata)
> >
> > > > print ' endposition : (%f, %f)' % (erelease.xdata,
> > > > erelease.ydata)
> > > > print ' used button : ', eclick.button
> > > >
> > > > def toggle_selector(event):
> > > > print ' Key pressed.'
> > > > if event.key in ['Q', 'q'] and toggle_selector.RS.active:
> > > > print ' RectangleSelector deactivated.'
> > > > toggle_selector.RS.set_active(False)
> > > > if event.key in ['A', 'a'] and not
> >
> > toggle_selector.RS.active:
> > > > print ' RectangleSelector activated.'
> > > > toggle_selector.RS.set_active(True)
> > > >
> > > > x = arange(100)/(99.0)
> > > > y = sin(x)
> > > > fig = figure
> > > > ax = subplot(111)
> > > > ax.plot(x,y)
> > > >
> > > > toggle_selector.RS = RectangleSelector(ax, onselect,
> > > > drawtype='line')
> > > > connect('key_press_event', toggle_selector)
> > > > show()
> > > > """
> > > >
> > > > On Friday 17 April 2009 02:26:51 Gökhan SEVER wrote:
> > > > > Hello,
> > > > >
> > > > > A quick question:
> > > > >
> > > > > I am using two numpy arrays to plot the figure shown in attachment.
> >
> > Is
> >
> > > > > it possible to get array indices of selected X-axes while using the
> > > > > zoom function? Later I can create a new figure from this selected
> > > > > portion instead of the same figure and/or apply an analysis.
> > > > >
> > > > > Thank you.
> >
> > -------------------------------------------------------------------------
> >
> > > >----- Stay on top of everything new and different, both inside and
> > > > around Java (TM) technology - register by April 22, and save
> > > > $200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco.
> > > > 300 plus technical and hands-on sessions. Register today.
> > > > Use priority code J9JMT32. http://p.sf.net/sfu/p
> > > > _______________________________________________
> > > > Matplotlib-users mailing list
> > > > Matplotlib-users@lists.sourceforge.net
> > > > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> >
> > -------------------------------------------------------------------------
> >----- Stay on top of everything new and different, both inside and
> > around Java (TM) technology - register by April 22, and save
> > $200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco.
> > 300 plus technical and hands-on sessions. Register today.
> > Use priority code J9JMT32. http://p.sf.net/sfu/p
> > _______________________________________________
> > Matplotlib-users mailing list
> > Matplotlib-users@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
rect.py
Description: application/python
------------------------------------------------------------------------------ Stay on top of everything new and different, both inside and around Java (TM) technology - register by April 22, and save $200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco. 300 plus technical and hands-on sessions. Register today. Use priority code J9JMT32. http://p.sf.net/sfu/p
_______________________________________________ Matplotlib-users mailing list Matplotlib-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/matplotlib-users
