* [#15284](http://bugzilla.scilab.org/show_bug.cgi?id=15284): Port names are not set to the corresponding I/O block labels.
* [#15310](http://bugzilla.scilab.org/show_bug.cgi?id=15310): `isdef` considered void arguments as defined (regression)
* [#15346](http://bugzilla.scilab.org/show_bug.cgi?id=15346): In an array of rationals, there was no way to address components with their linearized indices.
-* [#15403](http://bugzilla.scilab.org/show_bug.cgi?id=15403): `bar(..,"stacked")` could start from y<>0. Input arguments and possible syntaxes were poorly tested.
+* [#15403](http://bugzilla.scilab.org/show_bug.cgi?id=15403): `bar(..,"stacked")` could start from y<>0. Input arguments and possible syntaxes were poorly tested.
* [#15404](http://bugzilla.scilab.org/show_bug.cgi?id=15404): `surf()` and `mesh()` did not allow to specify `foreground`, `facecolor`, `markforeground` and `markbackground` global properties colors as a predefined named color out of a list of the 9 main color names. Colors specifications as "#RRGGBB" hexa code or Colors indices in the color map were nor allowed.
* [#15422](http://bugzilla.scilab.org/show_bug.cgi?id=15422): `strsubst("ab", "", "cd")` crashed Scilab.
* [#15423](http://bugzilla.scilab.org/show_bug.cgi?id=15423): `tbx_make(myModule,sections)` executed the existing builder (if any), instead of targeting only selected module sections. Otherwise, tbx_make(myModule, "help"|"macros") yielded an error, and tbx_make(myModule,"localization") never built it.
* [#15860](http://bugzilla.scilab.org/show_bug.cgi?id=15860): Horizontal concatenations with cblock tables had troubles.
* [#15866](http://bugzilla.scilab.org/show_bug.cgi?id=15866): Unlike all other set functions, `setdiff` did not accept any "r" or "c" option.
* [#15867](http://bugzilla.scilab.org/show_bug.cgi?id=15867): For input encoded integers, `setdiff` returned selected elements in decreasing order.
+* [#15873](http://bugzilla.scilab.org/show_bug.cgi?id=15873): In average, `sprand(100, 100, 0.8)` yielded ~8800 non-zero values instead of ~8000.
* [#15878](http://bugzilla.scilab.org/show_bug.cgi?id=15878): `sgrid` and `evans` were broken.
* [#15880](http://bugzilla.scilab.org/show_bug.cgi?id=15880): `sgrid` needed some improvements: Labeling was sometimes ambiguous ; large circles were not labeled ; data_bounds did not always take the input wn into account ; named and #RRGGBB colors specifications could not be used. `evans` needed some improvements: the block of legends hid data ; asymptotes were too visible.
* [#15881](http://bugzilla.scilab.org/show_bug.cgi?id=15881): Demos about stems, bars and histograms were spread in several subsections.
// Scilab ( http://www.scilab.org/ ) - This file is part of Scilab
// Copyright (C) INRIA
// Copyright (C) 2010 - DIGITEO - Michael Baudin
-//
// Copyright (C) 2012 - 2016 - Scilab Enterprises
+// Copyright (C) 2018 - Samuel GOUGEON
//
// This file is hereby licensed under the terms of the GNU GPL v2.0,
// pursuant to article 5.3.4 of the CeCILL v.2.1.
// For more information, see the COPYING file which you should have received
// along with this program.
-function a=sprand(m,n,density,typ)
+function a = sprand(m, n, density, law)
- //---- Check arguments----------------------------------------------
+ // CHECK ARGUMENTS
+ // ---------------
rhs = argn(2)
-
- if ( rhs < 3 | rhs > 4 ) then
+ if rhs < 3 | rhs > 4 then
error(msprintf(gettext("%s: Wrong number of input arguments: %d or %d expected.\n"), "sprand" , 3 , 4 ));
end
-
- if ( rhs < 4 ) then
- typ="def";
+ if rhs < 4 then
+ law = "unf";
end
-
- if type(typ)<>10 then
+ if m==0 | n==0 then
+ a = sparse([], [], [0 0])
+ return
+ end
+ // law
+ if type(law)<>10 then
error(msprintf(gettext("%s: Wrong type for input argument #%d: string expected.\n"),"sprand",4));
end
-
- if and(typ<>["u";"n";"uniform";"normal";"def";"nor"]) then
+ if and(law<>["u" "unf" "uniform" "def" "n" "nor" "normal"]) then
error(msprintf(gettext("%s: Wrong value for input argument #%d: ''%s'' or ''%s'' expected.\n"),"sprand",4,"uniform","normal"));
end
- if typ == "u" | typ == "uniform" then //"uniform" is the syntax for uniform distribution with rand, the equivalent with grand is "def"
- typ = "def";
- elseif typ == "n" | typ == "normal" then //"normal" is the syntax for normal distribution with rand, the equivalent with grand is "nor"
- typ = "nor";
+ if or(law == ["u" "uniform" "def"]) then
+ law = "unf";
+ elseif or(law == ["n" "normal"]) then
+ law = "nor";
end
- density=max(min(density,1),0);
-
- nel=m*n*density; //the objective number of non zero elements
+ // INITIALIZATION
+ // --------------
+ N = m*n
+ density = max(min(density,1),0);
+ nel = round(N*density); // the objective number of non zero elements
if nel==0 then
- a=sparse([],[],[m,n])
- return
- end
-
- //---- generate a sequence of increments----------------------------
- mdist = 1/density //the mean distance between to consecutive index
- nel1 = (2.2-density)*nel; //generate more increments than requested nnz elements
- if nel1<1
- ij = 1;
- else
- ij = round(1+grand(nel1,1,"exp",(mdist-1)))
- end
-
- //---- sum the increments to get the index--------------------------
- ij=cumsum(ij);
-
- //---- eliminate the index with exceed the maximum matrix index
- ij(find(ij>m*n))=[];
- nel1=size(ij,"*");
- if nel1==0 then
- a=sparse([],[],[m,n]);
+ a = sparse([],[], [m,n])
return
end
- //---- generate the row and column indices from the 1D index--------
-
- ij=ind2sub([m,n],ij);
-
- //---- generates the random non zeros elements --------------------
- //according to the requested law and create the sparse matrix
- if typ == "nor" then // Because of the syntax of grand, we have two cases, one with "def" and one with "nor"
- a=sparse(ij,grand(nel1,1,typ,0,1),[m,n]);
- elseif typ == "def" then
- a=sparse(ij,grand(nel1,1,typ),[m,n]);
- end
-
+ // GENERATION OF RANDOM UNIQUE INDICES OF NON-ZERO VALUES
+ // ------------------------------------------------------
+ // generate a sequence of increments greater than 1 ------------------------
+ di = N / nel
+ mdist = 1/density //the mean distance between two consecutive non-zero values
+ ij = grand(nel*1.01, 1, "exp", (mdist-1))
+
+ s = (nel + sum(ij(1:nel)))/(N-di/2) // rescaling factor
+ ij = 1 + ij/s // rescales intervals & guaranties that indices are unique
+ ij = floor(cumsum(ij)); // Intervals between indices => indices
+ ij(find(ij > N)) = []; // Garanties that linear indices are <= m*n
+ nel = length(ij)
+ ij = ind2sub([m,n], ij); // => [I J] conversion
+
+ // RESULT
+ // ------
+ a = sparse(ij, grand(nel, 1, law, 0, 1),[m,n]);
endfunction
+++ /dev/null
-// =============================================================================
-// Scilab ( http://www.scilab.org/ ) - This file is part of Scilab
-// Copyright (C) 2010-2011 - DIGITEO - Michael Baudin
-//
-// This file is distributed under the same license as the Scilab package.
-// =============================================================================
-// <-- CLI SHELL MODE -->
-// Check default "typ" = "uniform"
-rand("seed",0);
-grand("setsd",0);
-nrows = 1000;
-ncols = 2000;
-density = 1/100;
-s=sprand(nrows,ncols,density);
-assert_checkequal ( size(s) , [nrows,ncols] );
-nnzs=nnz(s);
-[ij,values]=spget(s);
-assert_checkequal ( min(values) >= 0 , %t );
-assert_checkequal ( max(values) <= 1 , %t );
-assert_checkalmostequal ( mean(values) , 0.5 , 1.e-2 );
-// Get empty matrix
-s=sprand(0,0,0.01);
-assert_checkequal ( s , [] );
-// Test the scientific part
-// In the following script, we check that the entries of the matrix have the expected distribution.
-// We use the spget function in order to get the nonzero entries.
-// Then we compute the min, mean and max of the entries and compare them with the limit values.
-rand("seed",0);
-grand("setsd",0);
-typ = "normal";
-nrows = 1000;
-ncols = 2000;
-density = 1/100;
-s=sprand(nrows,ncols,density,typ);
-assert_checkequal ( size(s) , [nrows,ncols] );
-nnzs=nnz(s);
-[ij,values]=spget(s);
-assert_checkequal ( nnzs > 10000 , %t );
-assert_checkequal ( min(values) < -3 , %t );
-assert_checkalmostequal ( mean(values) , 0 , [] , 1.e-2 );
-assert_checkequal ( max(values) > 3 , %t );
-assert_checkalmostequal ( variance(values) , 1 , 1.e-2 );
-rand("seed",0);
-grand("setsd",0);
-typ = "uniform";
-nrows = 1000;
-ncols = 2000;
-density = 1/100;
-s=sprand(nrows,ncols,density,typ);
-assert_checkequal ( size(s) , [nrows,ncols] );
-nnzs=nnz(s);
-[ij,values]=spget(s);
-assert_checkequal ( nnzs > 10000 , %t );
-assert_checkalmostequal ( min(values) , 0 , [] , 1.e-2 );
-assert_checkalmostequal ( mean(values) , 0.5 , 1.e-2 );
-assert_checkalmostequal ( max(values) , 1 , 1.e-2 );
-assert_checkalmostequal ( variance(values) , 1/12 , 1.e-2 );
-// In the following script, we check that the entry indices,
-// which are also chosen at random, have the correct distribution.
-// We generate kmax sparse random matrices with uniform distribution.
-// For each matrix, we consider the indices of the nonzero entries
-// which were generated, i.e. we see if the event Aij = {the entry (i,j) is nonzero}
-// occurred for each i and j, for i=1,2,...,nrows and j=1,2,...,ncols.
-// The matrix C(i,j) stores the number of times that the event Aij occurred.
-// The matrix R(k) stores the actual density of the try number k, where k=1,2,...,kmax.
-rand("seed",0);
-grand("setsd",0);
-kmax = 1000;
-ncols=100;
-nrows=100;
-density=0.01;
-typ="uniform";
-C=zeros(nrows,ncols);
-R=[];
-for k=1:kmax
- M=sprand(nrows,ncols,density,typ);
- NZ=find(M<>0);
- NZratio = size(NZ,"*")/(nrows*ncols);
- R=[R NZratio];
- C(NZ)=C(NZ)+1;
-end
-// Now that this algorithm has been performed (which may require some time),
-// we can compute elementary statistics to check that the algorithm performed well.
-// The average number should be close to the expectation.
-assert_checkalmostequal ( density*kmax , mean(C) , 1.e-2 );
-// The density should be close to expected density
-assert_checkalmostequal ( density , mean(R) , 1.e-2 );
-// More deeper tests should involve the particular distribution of
-// C, which follows a binomial law.
-// May be a chi-square test should be used for this.
// =============================================================================
// Scilab ( http://www.scilab.org/ ) - This file is part of Scilab
// Copyright (C) 2010-2011 - DIGITEO - Michael Baudin
+// Copyright (C) 2018 - Samuel GOUGEON
//
// This file is distributed under the same license as the Scilab package.
// =============================================================================
// <-- CLI SHELL MODE -->
+// <-- NO CHECK REF -->
-
+// Unitary tests of sprand()
+// =========================
// Check default "typ" = "uniform"
-rand("seed",0);
grand("setsd",0);
nrows = 1000;
ncols = 2000;
// Get empty matrix
s=sprand(0,0,0.01);
-assert_checkequal ( s , [] );
+assert_checkequal ( s , sparse([],[],[0,0]) );
// Test the scientific part
-// In the following script, we check that the entries of the matrix have the expected distribution.
-// We use the spget function in order to get the nonzero entries.
+// In the following script, we check that the entries of the matrix have the expected distribution.
+// We use the spget function in order to get the nonzero entries.
// Then we compute the min, mean and max of the entries and compare them with the limit values.
-rand("seed",0);
grand("setsd",0);
typ = "normal";
nrows = 1000;
assert_checkequal ( min(values) < -3 , %t );
assert_checkalmostequal ( mean(values) , 0 , [] , 1.e-2 );
assert_checkequal ( max(values) > 3 , %t );
-assert_checkalmostequal ( variance(values) , 1 , 1.e-2 );
+assert_checkalmostequal ( variance(values) , 1 , 2.2e-2 );
-rand("seed",0);
grand("setsd",0);
typ = "uniform";
nrows = 1000;
assert_checkalmostequal ( max(values) , 1 , 1.e-2 );
assert_checkalmostequal ( variance(values) , 1/12 , 1.e-2 );
-// In the following script, we check that the entry indices,
-// which are also chosen at random, have the correct distribution.
-// We generate kmax sparse random matrices with uniform distribution.
-// For each matrix, we consider the indices of the nonzero entries
+// In the following script, we check that the entry indices,
+// which are also chosen at random, have the correct distribution.
+// We generate kmax sparse random matrices with uniform distribution.
+// For each matrix, we consider the indices of the nonzero entries
// which were generated, i.e. we see if the event Aij = {the entry (i,j) is nonzero}
-// occurred for each i and j, for i=1,2,...,nrows and j=1,2,...,ncols.
-// The matrix C(i,j) stores the number of times that the event Aij occurred.
+// occurred for each i and j, for i=1,2,...,nrows and j=1,2,...,ncols.
+// The matrix C(i,j) stores the number of times that the event Aij occurred.
// The matrix R(k) stores the actual density of the try number k, where k=1,2,...,kmax.
-rand("seed",0);
grand("setsd",0);
kmax = 1000;
ncols=100;
R=[R NZratio];
C(NZ)=C(NZ)+1;
end
-// Now that this algorithm has been performed (which may require some time),
+// Now that this algorithm has been performed (which may require some time),
// we can compute elementary statistics to check that the algorithm performed well.
// The average number should be close to the expectation.
// The density should be close to expected density
assert_checkalmostequal ( density , mean(R) , 1.e-2 );
-// More deeper tests should involve the particular distribution of
+// More deeper tests should involve the particular distribution of
// C, which follows a binomial law.
// May be a chi-square test should be used for this.
-
+// CHECK THE ACTUAL DENSITY
+// ------------------------
+clc
+grand("setsd",0);
+for n = [100 1000 1e4 1e5 1e6]
+ for d = [1e-5 3e-5 1e-4 3e-4 1e-3 0.003 0.01 0.03 0.1 0.3 1]
+ clear m
+ for i = 1:min(5e6/n, 50)
+ s = sprand(1,n,d);
+ m(i) = nnz(s);
+ end
+ assert_checkalmostequal(mean(m), d*n, 1e-3, 2);
+ end
+end
projects / scilab.git / commitdiff