Related
I am new to Mathematica.
I have a lower triangular matrix defined as follow
A = Table[If[i > j, Subscript[a, i, j], 0], {i, s}, {j, s}];
I would like to the lower triangular elements in a list. For example, when s = 2, the list would contain listOfElement = {a_{2,1}} and for s = 3, listOfElement = {a_{2,1},a_{3,1},a_{3,2}}
How can I do this in Mathematica?
Thank you so much in advance
for example this
A = RandomReal[{0, 1}, {3, 3}];
MatrixForm[A]
M = First[Dimensions[A]];
Flatten[A[[# + 1 ;;, #]] & /# Range[M - 1]]
produces:
(0.586886 0.968229 0.543306
0.107212 0.0492116 0.103052
0.0569797 0.429895 0.70289
)
{0.107212,0.0569797,0.429895}
You can use Pick together with a selection matrix:
selectionMatrix = LowerTriangularize[ConstantArray[1, {s, s}], -1]
selectionMatrix is now a lower triangular matrix with ones where you want to Pick elements in A. You then get the elements of A like this:
listOfElements = Flatten # Pick[A, selectionMatrix, 1]
edit: Make sure you define s, of course.
I have written code which draws the Sierpinski fractal. It is really slow since it uses recursion. Do any of you know how I could write the same code without recursion in order for it to be quicker? Here is my code:
midpoint[p1_, p2_] := Mean[{p1, p2}]
trianglesurface[A_, B_, C_] := Graphics[Polygon[{A, B, C}]]
sierpinski[A_, B_, C_, 0] := trianglesurface[A, B, C]
sierpinski[A_, B_, C_, n_Integer] :=
Show[
sierpinski[A, midpoint[A, B], midpoint[C, A], n - 1],
sierpinski[B, midpoint[A, B], midpoint[B, C], n - 1],
sierpinski[C, midpoint[C, A], midpoint[C, B], n - 1]
]
edit:
I have written it with the Chaos Game approach in case someone is interested. Thank you for your great answers!
Here is the code:
random[A_, B_, C_] := Module[{a, result},
a = RandomInteger[2];
Which[a == 0, result = A,
a == 1, result = B,
a == 2, result = C]]
Chaos[A_List, B_List, C_List, S_List, n_Integer] :=
Module[{list},
list = NestList[Mean[{random[A, B, C], #}] &,
Mean[{random[A, B, C], S}], n];
ListPlot[list, Axes -> False, PlotStyle -> PointSize[0.001]]]
This uses Scale and Translate in combination with Nest to create the list of triangles.
Manipulate[
Graphics[{Nest[
Translate[Scale[#, 1/2, {0, 0}], pts/2] &, {Polygon[pts]}, depth]},
PlotRange -> {{0, 1}, {0, 1}}, PlotRangePadding -> .2],
{{pts, {{0, 0}, {1, 0}, {1/2, 1}}}, Locator},
{{depth, 4}, Range[7]}]
If you would like a high-quality approximation of the Sierpinski triangle, you can use an approach called the chaos game. The idea is as follows - pick three points that you wish to define as the vertices of the Sierpinski triangle and choose one of those points randomly. Then, repeat the following procedure as long as you'd like:
Choose a random vertex of the trangle.
Move from the current point to the halfway point between its current location and that vertex of the triangle.
Plot a pixel at that point.
As you can see at this animation, this procedure will eventually trace out a high-resolution version of the triangle. If you'd like, you can multithread it to have multiple processes plotting pixels at once, which will end up drawing the triangle more quickly.
Alternatively, if you just want to translate your recursive code into iterative code, one option would be to use a worklist approach. Maintain a stack (or queue) that contains a collection of records, each of which holds the vertices of the triangle and the number n. Initially put into this worklist the vertices of the main triangle and the fractal depth. Then:
While the worklist is not empty:
Remove the first element from the worklist.
If its n value is not zero:
Draw the triangle connecting the midpoints of the triangle.
For each subtriangle, add that triangle with n-value n - 1 to the worklist.
This essentially simulates the recursion iteratively.
Hope this helps!
You may try
l = {{{{0, 1}, {1, 0}, {0, 0}}, 8}};
g = {};
While [l != {},
k = l[[1, 1]];
n = l[[1, 2]];
l = Rest[l];
If[n != 0,
AppendTo[g, k];
(AppendTo[l, {{#1, Mean[{#1, #2}], Mean[{#1, #3}]}, n - 1}] & ## #) & /#
NestList[RotateLeft, k, 2]
]]
Show#Graphics[{EdgeForm[Thin], Pink,Polygon#g}]
And then replace the AppendTo by something more efficient. See for example https://mathematica.stackexchange.com/questions/845/internalbag-inside-compile
Edit
Faster:
f[1] = {{{0, 1}, {1, 0}, {0, 0}}, 8};
i = 1;
g = {};
While[i != 0,
k = f[i][[1]];
n = f[i][[2]];
i--;
If[n != 0,
g = Join[g, k];
{f[i + 1], f[i + 2], f[i + 3]} =
({{#1, Mean[{#1, #2}], Mean[{#1, #3}]}, n - 1} & ## #) & /#
NestList[RotateLeft, k, 2];
i = i + 3
]]
Show#Graphics[{EdgeForm[Thin], Pink, Polygon#g}]
Since the triangle-based functions have already been well covered, here is a raster based approach.
This iteratively constructs pascal's triangle, then takes modulo 2 and plots the result.
NestList[{0, ##} + {##, 0} & ## # &, {1}, 511] ~Mod~ 2 // ArrayPlot
Clear["`*"];
sierpinski[{a_, b_, c_}] :=
With[{ab = (a + b)/2, bc = (b + c)/2, ca = (a + c)/2},
{{a, ab, ca}, {ab, b, bc}, {ca, bc, c}}];
pts = {{0, 0}, {1, 0}, {1/2, Sqrt[3]/2}} // N;
n = 5;
d = Nest[Join ## sierpinski /# # &, {pts}, n]; // AbsoluteTiming
Graphics[{EdgeForm#Black, Polygon#d}]
(*sierpinski=Map[Mean, Tuples[#,2]~Partition~3 ,{2}]&;*)
Here is a 3D version,https://mathematica.stackexchange.com/questions/22256/how-can-i-compile-this-function
ListPlot#NestList[(# + RandomChoice[{{0, 0}, {2, 0}, {1, 2}}])/2 &,
N#{0, 0}, 10^4]
With[{data =
NestList[(# + RandomChoice#{{0, 0}, {1, 0}, {.5, .8}})/2 &,
N#{0, 0}, 10^4]},
Graphics[Point[data,
VertexColors -> ({1, #[[1]], #[[2]]} & /# Rescale#data)]]
]
With[{v = {{0, 0, 0.6}, {-0.3, -0.5, -0.2}, {-0.3, 0.5, -0.2}, {0.6,
0, -0.2}}},
ListPointPlot3D[
NestList[(# + RandomChoice[v])/2 &, N#{0, 0, 0}, 10^4],
BoxRatios -> 1, ColorFunction -> "Pastel"]
]
I am wondering if anyone can help me to plot the Cantor dust on the plane in Mathematica. This is linked to the Cantor set.
Thanks a lot.
EDIT
I actually wanted to have something like this:
Here's a naive and probably not very optimized way of reproducing the graphics for the ternary Cantor set construction:
cantorRule = Line[{{a_, n_}, {b_, n_}}] :>
With[{d = b - a, np = n - .1},
{Line[{{a, np}, {a + d/3, np}}], Line[{{b - d/3, np}, {b, np}}]}]
Graphics[{CapForm["Butt"], Thickness[.05],
Flatten#NestList[#/.cantorRule&, Line[{{0., 0}, {1., 0}}], 6]}]
To make Cantor dust using the same replacement rules, we take the result at a particular level, e.g. 4:
dust4=Flatten#Nest[#/.cantorRule&,Line[{{0.,0},{1.,0}}],4]/.Line[{{a_,_},{b_,_}}]:>{a,b}
and take tuples of it
dust4 = Transpose /# Tuples[dust4, 2];
Then we just plot the rectangles
Graphics[Rectangle ### dust4]
Edit: Cantor dust + squares
Changed specs -> New, but similar, solution (still not optimized).
Set n to be a positive integer and choice any subset of 1,...,n then
n = 3; choice = {1, 3};
CanDChoice = c:CanD[__]/;Length[c]===n :> CanD[c[[choice]]];
splitRange = {a_, b_} :> With[{d = (b - a + 0.)/n},
CanD##NestList[# + d &, {a, a + d}, n - 1]];
cantLevToRect[lev_]:=Rectangle###(Transpose/#Tuples[{lev}/.CanD->Sequence,2])
dust = NestList[# /. CanDChoice /. splitRange &, {0, 1}, 4] // Rest;
Graphics[{FaceForm[LightGray], EdgeForm[Black],
Table[cantLevToRect[lev], {lev, Most#dust}],
FaceForm[Black], cantLevToRect[Last#dust /. CanDChoice]}]
Here's the graphics for
n = 7; choice = {1, 2, 4, 6, 7};
dust = NestList[# /. CanDChoice /. splitRange &, {0, 1}, 2] // Rest;
and everything else the same:
Once can use the following approach. Define cantor function:
cantorF[r:(0|1)] = r;
cantorF[r_Rational /; 0 < r < 1] :=
Module[{digs, scale}, {digs, scale} = RealDigits[r, 3];
If[! FreeQ[digs, 1],
digs = Append[TakeWhile[Most[digs]~Join~Last[digs], # != 1 &], 1];];
FromDigits[{digs, scale}, 2]]
Then form the dust by computing differences of F[n/3^k]-F[(n+1/2)/3^k]:
With[{k = 4},
Outer[Times, #, #] &[
Table[(cantorF[(n + 1/2)/3^k] - cantorF[(n)/3^k]), {n, 0,
3^k - 1}]]] // ArrayPlot
I like recursive functions, so
cantor[size_, n_][pt_] :=
With[{s = size/3, ct = cantor[size/3, n - 1]},
{ct[pt], ct[pt + {2 s, 0}], ct[pt + {0, 2 s}], ct[pt + {2 s, 2 s}]}
]
cantor[size_, 0][pt_] := Rectangle[pt, pt + {size, size}]
drawCantor[n_] := Graphics[cantor[1, n][{0, 0}]]
drawCantor[5]
Explanation: size is the edge length of the square the set fits into. pt is the {x,y} coordinates of it lower left corner.
I have a function f(x,y) of two variables, of which I need to know the location of the curves at which it crosses zero. ContourPlot does that very efficiently (that is: it uses clever multi-grid methods, not just a brute force fine-grained scan) but just gives me a plot. I would like to have a set of values {x,y} (with some specified resolution) or perhaps some interpolating function which allows me to get access to the location of these contours.
Have thought of extracting this from the FullForm of ContourPlot but this seems to be a bit of a hack. Any better way to do this?
If you end up extracting points from ContourPlot, this is one easy way to do it:
points = Cases[
Normal#ContourPlot[Sin[x] Sin[y] == 1/2, {x, -3, 3}, {y, -3, 3}],
Line[pts_] -> pts,
Infinity
]
Join ## points (* if you don't want disjoint components to be separate *)
EDIT
It appears that ContourPlot does not produce very precise contours. They're of course meant for plotting and are good enough for that, but the points don't lie precisely on the contours:
In[78]:= Take[Join ## points /. {x_, y_} -> Sin[x] Sin[y] - 1/2, 10]
Out[78]= {0.000163608, 0.0000781187, 0.000522698, 0.000516078,
0.000282781, 0.000659909, 0.000626086, 0.0000917416, 0.000470424,
0.0000545409}
We can try to come up with our own method to trace the contour, but it's a lot of trouble to do it in a general way. Here's a concept that works for smoothly varying functions that have smooth contours:
Start from some point (pt0), and find the intersection with the contour along the gradient of f.
Now we have a point on the contour. Move along the tangent of the contour by a fixed step (resolution), then repeat from step 1.
Here's a basic implementation that only works with functions that can be differentiated symbolically:
rot90[{x_, y_}] := {y, -x}
step[f_, pt : {x_, y_}, pt0 : {x0_, y0_}, resolution_] :=
Module[
{grad, grad0, t, contourPoint},
grad = D[f, {pt}];
grad0 = grad /. Thread[pt -> pt0];
contourPoint =
grad0 t + pt0 /. First#FindRoot[f /. Thread[pt -> grad0 t + pt0], {t, 0}];
Sow[contourPoint];
grad = grad /. Thread[pt -> contourPoint];
contourPoint + rot90[grad] resolution
]
result = Reap[
NestList[step[Sin[x] Sin[y] - 1/2, {x, y}, #, .5] &, {1, 1}, 20]
];
ListPlot[{result[[1]], result[[-1, 1]]}, PlotStyle -> {Red, Black},
Joined -> True, AspectRatio -> Automatic, PlotMarkers -> Automatic]
The red points are the "starting points", while the black points are the trace of the contour.
EDIT 2
Perhaps it's an easier and better solution to use a similar technique to make the points that we get from ContourPlot more precise. Start from the initial point, then move along the gradient until we intersect the contour.
Note that this implementation will also work with functions that can't be differentiated symbolically. Just define the function as f[x_?NumericQ, y_?NumericQ] := ... if this is the case.
f[x_, y_] := Sin[x] Sin[y] - 1/2
refine[f_, pt0 : {x_, y_}] :=
Module[{grad, t},
grad = N[{Derivative[1, 0][f][x, y], Derivative[0, 1][f][x, y]}];
pt0 + grad*t /. FindRoot[f ## (pt0 + grad*t), {t, 0}]
]
points = Join ## Cases[
Normal#ContourPlot[f[x, y] == 0, {x, -3, 3}, {y, -3, 3}],
Line[pts_] -> pts,
Infinity
]
refine[f, #] & /# points
A slight variation for extracting points from ContourPlot (possibly due to David Park):
pts = Cases[
ContourPlot[Cos[x] + Cos[y] == 1/2, {x, 0, 4 Pi}, {y, 0, 4 Pi}],
x_GraphicsComplex :> First#x, Infinity];
or (as a list of {x,y} points)
ptsXY = Cases[
Cases[ContourPlot[
Cos[x] + Cos[y] == 1/2, {x, 0, 4 Pi}, {y, 0, 4 Pi}],
x_GraphicsComplex :> First#x, Infinity], {x_, y_}, Infinity];
Edit
As discussed here, an article by Paul Abbott in the Mathematica Journal (Finding Roots in an Interval) gives the following two alternative methods for obtaining a list of {x,y} values from ContourPlot, including (!)
ContourPlot[...][[1, 1]]
For the above example
ptsXY2 = ContourPlot[
Cos[x] + Cos[y] == 1/2, {x, 0, 4 Pi}, {y, 0, 4 Pi}][[1, 1]];
and
ptsXY3 = Cases[
Normal#ContourPlot[
Cos[x] + Cos[y] == 1/2, {x, 0, 4 Pi}, {y, 0, 4 Pi}],
Line[{x__}] :> x, Infinity];
where
ptsXY2 == ptsXY == ptsXY3
While looking at the belisarius's question about generation of non-singular integer matrices with uniform distribution of its elements, I was studying a paper by Dana Randal, "Efficient generation of random non-singular matrices". The algorithm proposed is recursive, and involves generating a matrix of lower dimension and assigning it to a given minor. I used combinations of Insert and Transpose to do it, but there are must be more efficient ways of doing it. How would you do it?
The following is the code:
Clear[Gen];
Gen[p_, 1] := {{{1}}, RandomInteger[{1, p - 1}, {1, 1}]};
Gen[p_, n_] := Module[{v, r, aa, tt, afr, am, tm},
While[True,
v = RandomInteger[{0, p - 1}, n];
r = LengthWhile[v, # == 0 &] + 1;
If[r <= n, Break[]]
];
afr = UnitVector[n, r];
{am, tm} = Gen[p, n - 1];
{Insert[
Transpose[
Insert[Transpose[am], RandomInteger[{0, p - 1}, n - 1], r]], afr,
1], Insert[
Transpose[Insert[Transpose[tm], ConstantArray[0, n - 1], r]], v,
r]}
]
NonSingularRandomMatrix[p_?PrimeQ, n_] := Mod[Dot ## Gen[p, n], p]
It does generate a non-singular matrix, and has uniform distribution of matrix elements, but requires p to be prime:
The code is also not every efficient, which is, I suspect due to my inefficient matrix constructors:
In[10]:= Timing[NonSingularRandomMatrix[101, 300];]
Out[10]= {0.421, Null}
EDIT So let me condense my question. The minor matrix of a given matrix m can be computed as follows:
MinorMatrix[m_?MatrixQ, {i_, j_}] :=
Drop[Transpose[Drop[Transpose[m], {j}]], {i}]
It is the original matrix with i-th row and j-th column deleted.
I now need to create a matrix of size n by n that will have the given minor matrix mm at position {i,j}. What I used in the algorithm was:
ExpandMinor[minmat_, {i_, j_}, v1_,
v2_] /; {Length[v1] - 1, Length[v2]} == Dimensions[minmat] :=
Insert[Transpose[Insert[Transpose[minmat], v2, j]], v1, i]
Example:
In[31]:= ExpandMinor[
IdentityMatrix[4], {2, 3}, {1, 2, 3, 4, 5}, {2, 3, 4, 4}]
Out[31]= {{1, 0, 2, 0, 0}, {1, 2, 3, 4, 5}, {0, 1, 3, 0, 0}, {0, 0, 4,
1, 0}, {0, 0, 4, 0, 1}}
I am hoping this can be done more efficiently, which is what I am soliciting in the question.
Per blisarius's suggestion I looked into implementing ExpandMinor via ArrayFlatten.
Clear[ExpandMinorAlt];
ExpandMinorAlt[m_, {i_ /; i > 1, j_}, v1_,
v2_] /; {Length[v1] - 1, Length[v2]} == Dimensions[m] :=
ArrayFlatten[{
{Part[m, ;; i - 1, ;; j - 1], Transpose#{v2[[;; i - 1]]},
Part[m, ;; i - 1, j ;;]},
{{v1[[;; j - 1]]}, {{v1[[j]]}}, {v1[[j + 1 ;;]]}},
{Part[m, i ;;, ;; j - 1], Transpose#{v2[[i ;;]]}, Part[m, i ;;, j ;;]}
}]
ExpandMinorAlt[m_, {1, j_}, v1_,
v2_] /; {Length[v1] - 1, Length[v2]} == Dimensions[m] :=
ArrayFlatten[{
{{v1[[;; j - 1]]}, {{v1[[j]]}}, {v1[[j + 1 ;;]]}},
{Part[m, All, ;; j - 1], Transpose#{v2}, Part[m, All, j ;;]}
}]
In[192]:= dim = 5;
mm = RandomInteger[{-5, 5}, {dim, dim}];
v1 = RandomInteger[{-5, 5}, dim + 1];
v2 = RandomInteger[{-5, 5}, dim];
In[196]:=
Table[ExpandMinor[mm, {i, j}, v1, v2] ==
ExpandMinorAlt[mm, {i, j}, v1, v2], {i, dim}, {j, dim}] //
Flatten // DeleteDuplicates
Out[196]= {True}
It took me a while to get here, but since I spent a good part of my postdoc generating random matrices, I could not help it, so here goes. The main inefficiency in the code comes from the necessity to move matrices around (copy them). If we could reformulate the algorithm so that we only modify a single matrix in place, we could win big. For this, we must compute the positions where the inserted vectors/rows will end up, given that we will typically insert in the middle of smaller matrices and thus shift the elements. This is possible. Here is the code:
gen = Compile[{{p, _Integer}, {n, _Integer}},
Module[{vmat = Table[0, {n}, {n}],
rs = Table[0, {n}],(* A vector of r-s*)
amatr = Table[0, {n}, {n}],
tmatr = Table[0, {n}, {n}],
i = 1,
v = Table[0, {n}],
r = n + 1,
rsc = Table[0, {n}], (* recomputed r-s *)
matstarts = Table[0, {n}], (* Horizontal positions of submatrix starts at a given step *)
remainingShifts = Table[0, {n}]
(*
** shifts that will be performed after a given row/vector insertion,
** and can affect the real positions where the elements will end up
*)
},
(*
** Compute the r-s and vectors v all at once. Pad smaller
** vectors v with zeros to fill a rectangular matrix
*)
For[i = 1, i <= n, i++,
While[True,
v = RandomInteger[{0, p - 1}, i];
For[r = 1, r <= i && v[[r]] == 0, r++];
If[r <= i,
vmat[[i]] = PadRight[v, n];
rs[[i]] = r;
Break[]]
]];
(*
** We must recompute the actual r-s, since the elements will
** move due to subsequent column insertions.
** The code below repeatedly adds shifts to the
** r-s on the left, resulting from insertions on the right.
** For example, if vector of r-s
** is {1,2,1,3}, it will become {1,2,1,3}->{2,3,1,3}->{2,4,1,3},
** and the end result shows where
** in the actual matrix the columns (and also rows for the case of
** tmatr) will be inserted
*)
rsc = rs;
For[i = 2, i <= n, i++,
remainingShifts = Take[rsc, i - 1];
For[r = 1, r <= i - 1, r++,
If[remainingShifts[[r]] == rsc[[i]],
Break[]
]
];
If[ r <= n,
rsc[[;; i - 1]] += UnitStep[rsc[[;; i - 1]] - rsc[[i]]]
]
];
(*
** Compute the starting left positions of sub-
** matrices at each step (1x1,2x2,etc)
*)
matstarts = FoldList[Min, First#rsc, Rest#rsc];
(* Initialize matrices - this replaces the recursion base *)
amatr[[n, rsc[[1]]]] = 1;
tmatr[[rsc[[1]], rsc[[1]]]] = RandomInteger[{1, p - 1}];
(* Repeatedly perform insertions - this replaces recursion *)
For[i = 2, i <= n, i++,
amatr[[n - i + 2 ;; n, rsc[[i]]]] = RandomInteger[{0, p - 1}, i - 1];
amatr[[n - i + 1, rsc[[i]]]] = 1;
tmatr[[n - i + 2 ;; n, rsc[[i]]]] = Table[0, {i - 1}];
tmatr[[rsc[[i]],
Fold[# + 1 - Unitize[# - #2] &,
matstarts[[i]] + Range[0, i - 1], Sort[Drop[rsc, i]]]]] =
vmat[[i, 1 ;; i]];
];
{amatr, tmatr}
],
{{FoldList[__], _Integer, 1}}, CompilationTarget -> "C"];
NonSignularRanomMatrix[p_?PrimeQ, n_] := Mod[Dot ## Gen[p, n],p];
NonSignularRanomMatrixAlt[p_?PrimeQ, n_] := Mod[Dot ## gen[p, n],p];
Here is the timing for the large matrix:
In[1114]:= gen [101, 300]; // Timing
Out[1114]= {0.078, Null}
For the histogram, I get the identical plots, and the 10-fold efficiency boost:
In[1118]:=
Histogram[Table[NonSignularRanomMatrix[11, 5][[2, 3]], {10^4}]]; // Timing
Out[1118]= {7.75, Null}
In[1119]:=
Histogram[Table[NonSignularRanomMatrixAlt[11, 5][[2, 3]], {10^4}]]; // Timing
Out[1119]= {0.687, Null}
I expect that upon careful profiling of the above compiled code, one could further improve the performance. Also, I did not use runtime Listable attribute in Compile, while this should be possible. It may also be that the parts of the code which perform assignment to minors are generic enough so that the logic can be factored out of the main function - I did not investigate that yet.
For the first part of your question (which I hope I understand properly) can
MinorMatrix be written as follows?
MinorMatrixAlt[m_?MatrixQ, {i_, j_}] := Drop[mat, {i}, {j}]