Single precision matrix multiplication

조회 수: 19 (최근 30일)
Donald Liu
Donald Liu 2019년 12월 12일
편집: James Tursa 2022년 8월 6일
Script:
M = 10000; N = 100;
A = round(10000 * complex(randn(M, N), randn(M, N)));
B = single(A') * single(A);
C = B' - B;
max(abs(C(:))) %// ==> answer is non-zero
A = single(A);
B = A' * A;
C = B' - B;
max(abs(C(:))) %// ==> answer is zero
Not sure what caused the difference? Thanks!
  댓글 수: 5
Bruno Luong
Bruno Luong 2022년 8월 5일
This test seems to show that there is no extra intelligent parsing, but simply the expression (C*D) returns numerical Hermitian when C==D' in R2022a.
M = 10000; N = 100;
A = round(10000 * complex(randn(M, N), randn(M, N)));
A = single(A);
SAc = single(A');
isequal(SAc, A')
ans = logical
1
tic
B1 = SAc * A;
toc
Elapsed time is 0.011465 seconds.
norm(B1-B1','fro')
ans = single 0
tic
B2 = A' * A;
toc
Elapsed time is 0.006590 seconds.
norm(B2-B2','fro')
ans = single 0
norm(B1-B2,'fro')/norm(B2)
ans = single 4.6414e-07
James Tursa
James Tursa 2022년 8월 6일
편집: James Tursa 2022년 8월 6일
I can only guess, then, that the underlying BLAS library algorithms have changed. The timings and B1-B2 differences confirm that two different BLAS subroutines are called as has been the case historically. I don't have the latest version installed to investigate this further.

댓글을 달려면 로그인하십시오.

채택된 답변

Matt J
Matt J 2019년 12월 12일
편집: Matt J 2019년 12월 12일
The Matlab interpreter runs a special dedicated "ctranspose-multiply" function to evaluate expressions of the form,
P'*Q
It does not do any explicit transposition of P, thus saving time and memory. This special function likely does a pre-check for the special case when P and Q are the same variable. In this case, it knows that it only has to compute the lower (or upper) triangular part of the resulting matrix, and can just copy the conjugate of that to the upper(lower) triangle. Thus, the output of A'*A will always be perfectly Hermitian.
However, for the expression,
single(P')*single(Q)
no special functions are used. Every operation in this expression (single,ctranpose,mtimes) is done separately and explicitly, and can generate floating point errors that may not be Hermitian across the final matrix.
  댓글 수: 2
James Tursa
James Tursa 2019년 12월 12일
편집: James Tursa 2019년 12월 12일
Yes, but "A'*A will always be perfectly symmetric" should instead read "A'*A will always be perfectly Hermitian". And the triangular copy is a conjugate copy, not a straight copy.
Matt J
Matt J 2019년 12월 12일
Right, you are! Fixed.

댓글을 달려면 로그인하십시오.

추가 답변 (2개)

James Tursa
James Tursa 2019년 12월 12일
편집: James Tursa 2019년 12월 12일
To illustrate what Matt is saying, a simple timing test:
>> format longg
>> S = round(10000*single(rand(5000)+rand(5000)*1i));
>> St = S';
>> tic;SaS=S'*S;toc
Elapsed time is 1.675010 seconds.
>> tic;StS=St*S;toc
Elapsed time is 2.942037 seconds.
>> isequal(SaS,StS)
ans =
logical
0
>> isequal(SaS',SaS)
ans =
logical
1
>> isequal(StS',StS)
ans =
logical
0
The difference in timing is because the S'*S method only does about half the multiplication work with some added conjugate copying. And since the off-diagonal elements are the result of a copy, you get an exact Hermitian result for this case. Not so for the generic multiply case ... more on that below.
Another thing to note is that even though the individual elements are made of up integers, the intermediate sums overflow the precision of a single precision variable. Hence the difference in the results depending on the order of calculations even though the result is also made up of exact integers. E.g.,
>> SaS(1,1)
ans =
single
3.353673e+11
>> StS(1,1)
ans =
single
3.353673e+11 + 3408i
>> dot(S(:,1),S(:,1))
ans =
single
3.353671e+11
>> S(:,1)'*S(:,1)
ans =
single
3.353672e+11
Four different answers for the (1,1) spot depending on how we do the calculation.
Now lower the integer values so the intermediate sums do not overflow the precision of a single precision variable:
>> S = round(10*single(rand(5000)+rand(5000)*1i));
>> St = S';
>> tic;SaS=S'*S;toc
Elapsed time is 1.807634 seconds.
>> tic;StS=St*S;toc
Elapsed time is 2.977281 seconds.
>> isequal(SaS,StS)
ans =
logical
1
>> SaS(1,1)
ans =
single
329958
>> StS(1,1)
ans =
single
329958
>> dot(S(:,1),S(:,1))
ans =
single
329958
>> S(:,1)'*S(:,1)
ans =
single
329958
Here we get exactly the same results for the four different methods because the sums didn't overflow the precision of single precision.
  댓글 수: 7
James Tursa
James Tursa 2022년 8월 5일
편집: James Tursa 2022년 8월 5일
If speed was not an issue, you could always write your own C mex routine to do the matrix multiply as a series of dot products using the BLAS dsdot( ) routine, which computes the dot product of single precision inputs using a double precision accumulator. This would be very easy to code and could even be multi-threaded, but would not run as fast as the single precision matrix multiply routines.
Walter Roberson
Walter Roberson 2022년 8월 5일
Furthermore, that single precision hermitian inner products are slower than
double precision ones means there is no (ok, extremely extremely rarely?
a) point in ever taking a single precision hermitian inner product.
S = round(10*single(rand(5000)+rand(5000)*1i));
St = S';
tic;SaS=S'*S;toc
Elapsed time is 1.596208 seconds.
tic;StS=St*S;toc
Elapsed time is 3.040935 seconds.
dS = double(S);
dSt = double(St);
tic; dSaS = dS'*dS;toc
Elapsed time is 3.155149 seconds.
tic; dStS = dSt*dS;toc
Elapsed time is 5.954180 seconds.
Double precision looks slower ?

댓글을 달려면 로그인하십시오.


Donald Liu
Donald Liu 2019년 12월 12일
After reading Matt's answer, I realized the following:
In single(A), we're converting element a + j * b to single precision a1 + j * b1.
In single(A'), the same element becomes a + j * (-b) and it is converted to single precision yielding a2 + j * (-b2).
Due to rounding differences of positive and negative numbers, b1 and b2 may not be equal. Thus single(A') * single(A) is not strictly Hermitian.
However, even single(A)' * single(A) is not strictly Hermitian, which is still puzzling.
Thanks!
  댓글 수: 4
James Tursa
James Tursa 2019년 12월 12일
You do the rounding before you do the multiply, so all the downstream calculations start with exactly the same values. The rounding you did has no effect on this. If everything starts with integral values, it won't matter what order you do the calculations AS LONG AS you don't overflow the precision of the variable type you are using. But in your case you did overflow this, hence the differences.
Matt J
Matt J 2019년 12월 12일
편집: Matt J 2019년 12월 12일
However, even single(A)' * single(A) is not strictly Hermitian, which is still puzzling.
Note that the first single(A) in the expression will not occupy the same memory address as the second single(A). Therefore, Matlab's ctranpose-multiply routine might not be recognizing them as the same matrix. Example:
>> format debug; A=rand(3); B=single(A), C=single(A),
B =
3×3 single matrix
Structure address = 13f3b4f00 %<-----Memory Address
m = 3
n = 3
pr = 12ebdd7e0
0.7265 0.2352 0.7033
0.6667 0.6863 0.5338
0.0327 0.2811 0.3319
C =
3×3 single matrix
Structure address = 13f37b270 %<-----Memory Address
m = 3
n = 3
pr = 12eb5dae0
0.7265 0.2352 0.7033
0.6667 0.6863 0.5338
0.0327 0.2811 0.3319
By contrast, in the following, Q and A do occupy the same memory address
M = 10000; N = 100;
A = round(10000 * complex(randn(M, N), randn(M, N)));
A = single(A);
Q=A;
B = Q' * A;
C = B' - B;
max(abs(C(:)))
ans =
single
0

댓글을 달려면 로그인하십시오.

카테고리

Help CenterFile Exchange에서 Logical에 대해 자세히 알아보기

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by