Menu
▾
▴
matplotlib-users
|
From: Matthew C. <mat...@gm...> - 2014-08-28 13:02:29
|
Hi Matplotlib Users! I have some 2-d arrays, which i am displaying with implot, and deriving contours for with contour. Easy - I'm just pulling them out of collections[0].get_paths() . However what's not easy is that I would like to recover a 1-0 or True-False array of the array values (pixels) that fall within the contours. Some line crossing algorithm/floodfill could do it, but I guess that matplotlib's fill() or contourf() must do this under the hood anyway. I've looked into the output both functions, but I don't see anything obvious.. Does anybody know if there's an a way to pull out a such an array from matplotlib? Any pointers are appreciated! Cheers, Matt |
|
From: Shantha K. <shk...@in...> - 2014-08-28 13:27:10
Attachments:
pic19302.gif
|
Hi All,
Thank you so much for your help, It really worked for me.
I need one more favor,
I have ploted the graph with 2 Y-axes
Here is the code
lns1 = ax1.plot(x1, y1, 'r-o',label=LY1,markersize=4)
ax1 = self.set_ylim(ax1,y1,label=LY1)
lns2 = ax2.plot(x2, y2, 'b-o',label=LY2,markersize=4)
I want to set the grid to the second y-axes in the same graph.
Please help on the same.
See here the difference
(Embedded image moved to file: pic19302.gif)
SHANTHA KUMARA REVANASIDDAPPA
Python Developper
Operations & Production Control, A1 Telekom
IBM
M +43-6642196132
@ Sha...@ex... |
|
From: Benjamin R. <ben...@ou...> - 2014-08-28 13:33:54
|
You are asking for the twinx() feature: http://matplotlib.org/examples/api/two_scales.html On Thu, Aug 28, 2014 at 9:24 AM, Shantha Kumara <shk...@in...> wrote: > Hi All, > > Thank you so much for your help, It really worked for me. > > > I need one more favor, > > > I have ploted the graph with 2 Y-axes > > Here is the code > > lns1 = ax1.plot(x1, y1, 'r-o',label=LY1,markersize=4) > ax1 = self.set_ylim(ax1,y1,label=LY1) > lns2 = ax2.plot(x2, y2, 'b-o',label=LY2,markersize=4) > > I want to set the grid to the second y-axes in the same graph. > > Please help on the same. > > See here the difference > > (Embedded image moved to file: pic19302.gif) > > > > SHANTHA KUMARA REVANASIDDAPPA > Python Developper > Operations & Production Control, A1 Telekom > IBM > M +43-6642196132 > @ Sha...@ex... > > ------------------------------------------------------------------------------ > Slashdot TV. > Video for Nerds. Stuff that matters. > http://tv.slashdot.org/ > _______________________________________________ > Matplotlib-users mailing list > Mat...@li... > https://lists.sourceforge.net/lists/listinfo/matplotlib-users > > |
|
From: Benjamin R. <ben...@ou...> - 2014-08-28 13:32:28
|
That stuff is done in the deep underbelly of matplotlib and isn't exposed to the user. It is done as part of the rendering process in AGG or whichever other backend is performing the render. I have done something very similar to what you are asking for my job, and while I can't share the code, I can point out that GDAL has a very fast polygon rasterizer. I can also point you to this link: http://stackoverflow.com/questions/2220749/rasterizing-a-gdal-layer I will also say that there are some subtle errors in that code, but it should get you to where you need to go. Cheers! Ben Root On Thu, Aug 28, 2014 at 9:02 AM, Matthew Czesarski < mat...@gm...> wrote: > Hi Matplotlib Users! > > > > I have some 2-d arrays, which i am displaying with implot, and deriving > contours for with contour. Easy - I'm just pulling them out of > collections[0].get_paths() . > > However what's not easy is that I would like to recover a 1-0 or > True-False array of the array values (pixels) that fall within the > contours. Some line crossing algorithm/floodfill could do it, but I guess > that matplotlib's fill() or contourf() must do this under the hood anyway. > I've looked into the output both functions, but I don't see anything > obvious.. > > Does anybody know if there's an a way to pull out a such an array from > matplotlib? Any pointers are appreciated! > > Cheers, > Matt > > > ------------------------------------------------------------------------------ > Slashdot TV. > Video for Nerds. Stuff that matters. > http://tv.slashdot.org/ > _______________________________________________ > Matplotlib-users mailing list > Mat...@li... > https://lists.sourceforge.net/lists/listinfo/matplotlib-users > > |
|
From: Eric F. <ef...@ha...> - 2014-08-29 01:23:30
|
On 2014/08/28, 3:02 AM, Matthew Czesarski wrote: > Hi Matplotlib Users! > > > > I have some 2-d arrays, which i am displaying with implot, and deriving > contours for with contour. Easy - I'm just pulling them out of > collections[0].get_paths() . > > However what's not easy is that I would like to recover a 1-0 or > True-False array of the array values (pixels) that fall within the > contours. Some line crossing algorithm/floodfill could do it, but I > guess that matplotlib's fill() or contourf() must do this under the hood > anyway. I've looked into the output both functions, but I don't see > anything obvious.. > > Does anybody know if there's an a way to pull out a such an array from > matplotlib? Any pointers are appreciated! Make an array of (x, y) pairs from the X and Y you use in your call to contour, and then feed that array to the contains_points() method of your contour Path. This will give you the desired Boolean array for any given Path; depending on what you want, you might need to combine arrays for more than one Path. To get closed paths, I think you will want to use contourf, not contour. Eric > > Cheers, > Matt > > > ------------------------------------------------------------------------------ > Slashdot TV. > Video for Nerds. Stuff that matters. > http://tv.slashdot.org/ > > > > _______________________________________________ > Matplotlib-users mailing list > Mat...@li... > https://lists.sourceforge.net/lists/listinfo/matplotlib-users > |
|
From: Joe K. <jof...@gm...> - 2014-08-29 02:18:27
|
Why not just use boolean indexing?
E.g. to find the region that falls between 5 and 10, do "(z >=5) & (z <=
10)":
In [1]: import numpy as np
In [2]: x, y = np.mgrid[-10:10, -10:10]
In [3]: z = np.hypot(x, y)
In [4]: result = (z >= 5) & (z <= 10)
In [5]: result.astype(int)
Out[5]:
array([[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0],
[0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0],
[0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0],
[0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0],
[0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0],
[0, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1],
[0, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1],
[0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1],
[0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1],
[1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1],
[0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1],
[0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1],
[0, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1],
[0, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1],
[0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0],
[0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0],
[0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0],
[0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0],
[0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0]])
Cheers,
-Joe
On Thu, Aug 28, 2014 at 8:23 PM, Eric Firing <ef...@ha...> wrote:
> On 2014/08/28, 3:02 AM, Matthew Czesarski wrote:
> > Hi Matplotlib Users!
> >
> >
> >
> > I have some 2-d arrays, which i am displaying with implot, and deriving
> > contours for with contour. Easy - I'm just pulling them out of
> > collections[0].get_paths() .
> >
> > However what's not easy is that I would like to recover a 1-0 or
> > True-False array of the array values (pixels) that fall within the
> > contours. Some line crossing algorithm/floodfill could do it, but I
> > guess that matplotlib's fill() or contourf() must do this under the hood
> > anyway. I've looked into the output both functions, but I don't see
> > anything obvious..
> >
> > Does anybody know if there's an a way to pull out a such an array from
> > matplotlib? Any pointers are appreciated!
>
> Make an array of (x, y) pairs from the X and Y you use in your call to
> contour, and then feed that array to the contains_points() method of
> your contour Path. This will give you the desired Boolean array for any
> given Path; depending on what you want, you might need to combine arrays
> for more than one Path.
>
> To get closed paths, I think you will want to use contourf, not contour.
>
> Eric
>
>
>
> >
> > Cheers,
> > Matt
> >
> >
> >
> ------------------------------------------------------------------------------
> > Slashdot TV.
> > Video for Nerds. Stuff that matters.
> > http://tv.slashdot.org/
> >
> >
> >
> > _______________________________________________
> > Matplotlib-users mailing list
> > Mat...@li...
> > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> >
>
>
>
> ------------------------------------------------------------------------------
> Slashdot TV.
> Video for Nerds. Stuff that matters.
> http://tv.slashdot.org/
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
|
|
From: Sterling S. <sm...@fu...> - 2014-08-29 05:19:03
|
Joe and list, This is off topic, but can you point me to good documentation on the use of '&' as opposed to numpy.logical_and ? Thanks, Sterling On Aug 28, 2014, at 7:18PM, Joe Kington wrote: > Why not just use boolean indexing? > > E.g. to find the region that falls between 5 and 10, do "(z >=5) & (z <= 10)": > > In [1]: import numpy as np > > In [2]: x, y = np.mgrid[-10:10, -10:10] > > In [3]: z = np.hypot(x, y) > > In [4]: result = (z >= 5) & (z <= 10) > > In [5]: result.astype(int) > Out[5]: > array([[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0], > [0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0], > [0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0], > [0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0], > [0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0], > [0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0], > [0, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1], > [0, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1], > [0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1], > [0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1], > [1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1], > [0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1], > [0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1], > [0, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1], > [0, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1], > [0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0], > [0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0], > [0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0], > [0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0], > [0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0]]) > > Cheers, > -Joe > > > > On Thu, Aug 28, 2014 at 8:23 PM, Eric Firing <ef...@ha...> wrote: > On 2014/08/28, 3:02 AM, Matthew Czesarski wrote: > > Hi Matplotlib Users! > > > > > > > > I have some 2-d arrays, which i am displaying with implot, and deriving > > contours for with contour. Easy - I'm just pulling them out of > > collections[0].get_paths() . > > > > However what's not easy is that I would like to recover a 1-0 or > > True-False array of the array values (pixels) that fall within the > > contours. Some line crossing algorithm/floodfill could do it, but I > > guess that matplotlib's fill() or contourf() must do this under the hood > > anyway. I've looked into the output both functions, but I don't see > > anything obvious.. > > > > Does anybody know if there's an a way to pull out a such an array from > > matplotlib? Any pointers are appreciated! > > Make an array of (x, y) pairs from the X and Y you use in your call to > contour, and then feed that array to the contains_points() method of > your contour Path. This will give you the desired Boolean array for any > given Path; depending on what you want, you might need to combine arrays > for more than one Path. > > To get closed paths, I think you will want to use contourf, not contour. > > Eric > > > > > > > Cheers, > > Matt > > > > > > ------------------------------------------------------------------------------ > > Slashdot TV. > > Video for Nerds. Stuff that matters. > > http://tv.slashdot.org/ > > > > > > > > _______________________________________________ > > Matplotlib-users mailing list > > Mat...@li... > > https://lists.sourceforge.net/lists/listinfo/matplotlib-users > > > > > ------------------------------------------------------------------------------ > Slashdot TV. > Video for Nerds. Stuff that matters. > http://tv.slashdot.org/ > _______________________________________________ > Matplotlib-users mailing list > Mat...@li... > https://lists.sourceforge.net/lists/listinfo/matplotlib-users > > ------------------------------------------------------------------------------ > Slashdot TV. > Video for Nerds. Stuff that matters. > http://tv.slashdot.org/_______________________________________________ > Matplotlib-users mailing list > Mat...@li... > https://lists.sourceforge.net/lists/listinfo/matplotlib-users |
|
From: Benjamin R. <ben...@ou...> - 2014-08-29 13:28:43
|
slaps forehead... Joe, you just won the "duh!" moment of the month award! Cheers! Ben Root On Thu, Aug 28, 2014 at 10:18 PM, Joe Kington <jof...@gm...> wrote: > Why not just use boolean indexing? > > E.g. to find the region that falls between 5 and 10, do "(z >=5) & (z <= > 10)": > > In [1]: import numpy as np > > In [2]: x, y = np.mgrid[-10:10, -10:10] > > In [3]: z = np.hypot(x, y) > > In [4]: result = (z >= 5) & (z <= 10) > > In [5]: result.astype(int) > Out[5]: > array([[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0], > [0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0], > [0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0], > [0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0], > [0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0], > [0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0], > [0, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1], > [0, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1], > [0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1], > [0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1], > [1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1], > [0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1], > [0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1], > [0, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1], > [0, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1], > [0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0], > [0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0], > [0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0], > [0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0], > [0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0]]) > > Cheers, > -Joe > > > > On Thu, Aug 28, 2014 at 8:23 PM, Eric Firing <ef...@ha...> wrote: > >> On 2014/08/28, 3:02 AM, Matthew Czesarski wrote: >> > Hi Matplotlib Users! >> > >> > >> > >> > I have some 2-d arrays, which i am displaying with implot, and deriving >> > contours for with contour. Easy - I'm just pulling them out of >> > collections[0].get_paths() . >> > >> > However what's not easy is that I would like to recover a 1-0 or >> > True-False array of the array values (pixels) that fall within the >> > contours. Some line crossing algorithm/floodfill could do it, but I >> > guess that matplotlib's fill() or contourf() must do this under the hood >> > anyway. I've looked into the output both functions, but I don't see >> > anything obvious.. >> > >> > Does anybody know if there's an a way to pull out a such an array from >> > matplotlib? Any pointers are appreciated! >> >> Make an array of (x, y) pairs from the X and Y you use in your call to >> contour, and then feed that array to the contains_points() method of >> your contour Path. This will give you the desired Boolean array for any >> given Path; depending on what you want, you might need to combine arrays >> for more than one Path. >> >> To get closed paths, I think you will want to use contourf, not contour. >> >> Eric >> >> >> >> > >> > Cheers, >> > Matt >> > >> > >> > >> ------------------------------------------------------------------------------ >> > Slashdot TV. >> > Video for Nerds. Stuff that matters. >> > http://tv.slashdot.org/ >> > >> > >> > >> > _______________________________________________ >> > Matplotlib-users mailing list >> > Mat...@li... >> > https://lists.sourceforge.net/lists/listinfo/matplotlib-users >> > >> >> >> >> ------------------------------------------------------------------------------ >> Slashdot TV. >> Video for Nerds. Stuff that matters. >> http://tv.slashdot.org/ >> _______________________________________________ >> Matplotlib-users mailing list >> Mat...@li... >> https://lists.sourceforge.net/lists/listinfo/matplotlib-users >> > > > > ------------------------------------------------------------------------------ > Slashdot TV. > Video for Nerds. Stuff that matters. > http://tv.slashdot.org/ > _______________________________________________ > Matplotlib-users mailing list > Mat...@li... > https://lists.sourceforge.net/lists/listinfo/matplotlib-users > > |
