Le 10/09/2018 à 12:57, [email protected] a écrit :
Le 10/09/2018 à 11:00, Stéphane Mottelet a écrit :
Le 10/09/2018 à 09:56, [email protected] a écrit :
Hi all,
I'm trying to plot the density of scatter points in (x,y).
Basically, I have a huge Xvec and a huge Yvec of same size that
represents the position of a huge number of points in a 2D plane.
I would like to get a density plot, that is given a certain grid, a
2D plot representing the number of scatter points within a each cell
of the grid.
In matplotib, this corresponds to functions like "hist2d" and "hexbin".
Is there any equivalent in Scilab?
Or do I have to calculate by hand my density matrix and then use
graplot or contourfill to plot my density?
Any pointer at the least inefficient way to do this calculation is
more than welcome...
Thank you in advance for your help,
Antoine
Hello Antoine,
The most efficient at the Scilab level would be to use the
incremental feature of "sparse":
function d=bins2d(x,y,xbins,ybins)
n = size(x,"*");
[ix,cx] = dsearch(x,xbins);
[iy,cy] = dsearch(y,ybins);
kin = find((ix>0) & (iy>0));
d = full(sparse([ix(kin) iy(kin)], ones(kin), [size(xbins,"*")-1,
size(ybins,"*")-1]));
end
x = rand(10000,1,'normal');
y = rand(10000,1,'normal');
--> d = bins2d(x,y,-5:5,-5:5)
d =
0. 0. 0. 0. 1. 0. 0. 0. 0. 0.
0. 0. 0. 3. 4. 4. 0. 0. 0. 0.
0. 0. 6. 31. 82. 80. 31. 2. 0. 0.
0. 0. 24. 183. 477. 477. 200. 25. 3. 0.
0. 5. 68. 439. 1131. 1158. 426. 57. 4. 0.
0. 5. 91. 485. 1168. 1220. 438. 72. 3. 0.
0. 4. 30. 190. 470. 472. 173. 21. 3. 0.
0. 0. 6. 30. 83. 78. 24. 2. 0. 0.
0. 1. 0. 2. 2. 3. 3. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
--> sum(d)
ans =
10000.
Thanks a lot, it works and is really fast even 2*10⁶ scatter points!
This feature is fixed in scilab-branch-6.0:
Huh, what do you mean? Is there a new function bins2d or hist2d in
scilab-branch-6.0?
No. "sparse" is broken in scilab-6.0.1, but fixed in scilab-branch-6.0.
Hence the above code will work in scilab-5.5.2 or >= branch-6.0
S.
Antoine
S.
--
Stéphane Mottelet
Ingénieur de recherche
EA 4297 Transformations Intégrées de la Matière Renouvelable
Département Génie des Procédés Industriels
Sorbonne Universités - Université de Technologie de Compiègne
CS 60319, 60203 Compiègne cedex
Tel : +33(0)344234688
http://www.utc.fr/~mottelet
_______________________________________________
users mailing list
[email protected]
http://lists.scilab.org/mailman/listinfo/users
