0

I have 2 arrays (for the sake of the example, let's name them A and B) and i perform the following manipulations at them, but i get an error at the assignment of "d2" in my code. 

n = len(tracks) #tracks is a list containing different-length 3d arrays 
n=30; #test with a few tracks 
length = len(tracks) #list containing the total number of "samples"
perm_index = np.random.permutation(length) #uniform sampling without replacement
subset_len = 5 # choose the size of subset of tracks A
subset_A = [tracks[x:x+1] for x in xrange(0, subset_len, 1)]
subset_B = [tracks[x:x+1] for x in xrange(subset_len, n, 1)]
tempA = distance_calc.dist_calcsub(len(subset_A), subset_A)  # distance matrix calculation
tempA = mcp.sym_mcp(len(subset_A), tempA)    # symmetrize mcp ???
tempB = distance_calc.dist_calcsubs(subset_A, subset_B)  # distance matrix calculation
#symmetrize mcp ? ? its not diagonal, symmetric . . .

A = affinity.aff_conv(60, tempA) # conversion to affinity 
B = affinity.aff_conv(60, tempB) # conversion to affinity 

#((row,col)) = np.shape(A)
#A = normalization_affinity.norm_aff(row, col, A) # normalization of affinity matrix

# Normalize A and B for Laplacian using row sums of W, where W = [A B; B' B'*A^-1*B].
# Let d1 = [A B]*1, d2 = [B' B'*A^-1*B]*1, dhat = sqrt(1./[d1; d2]).
d1 = np.sum( np.vstack((A, np.transpose(B))) )
d2 = np.sum(B,0) + np.dot(np.sum(np.transpose(B),0), np.dot(np.linalg.pinv(A), B ))
dhat = np.transpose(np.sqrt( 1/ np.hstack((d1, d2)) ))
A = A* np.dot( dhat[0:subset_len], np.transpose(dhat[0:subset_len]) )
B = B* np.dot( dhat[0:subset_len], np.transpose(dhat[subset_len:n]) )

The error again is "ValueError: matrices are not aligned." because the np.dot vectors are 1d vectors of different size; I know the reason why this is happening but I am following exactly the equations to perform the Nystrom method.

P.S: I am following the method described in p.90-92 in this thesis: thesis link

Improve this question
3
  • What shape did you expect np.sum(B, 0) to have? You're summing each row into a single value, so you get 10 values. Did you want a 10x4 matrix with each row replaced by 4 copies of the row's sum or something? Commented Oct 29, 2014 at 19:09
  • i am expecting to get the row sums of B in any case; The A and B are only for illustration purposes. In this case, i am getting exactly what i am expecting to get. Commented Oct 29, 2014 at 19:55
  • You really need to summarize the relevant information from that linked thesis so that the question makes sense on its own. Commented Oct 30, 2014 at 19:33

2 Answers 2

Reset to default
1

Looking at the paper, you've got two problems here.

Let's start with the information you left out of your question. You're trying to do this operation:

bc + B.T * A^−1 * br

where ar and br are column vectors containing the row sums of A and B and bc is the column sum of B.

In particular, you're mapping that A^-1 * br to np.dot( np.linalg.pinv(A), np.sum(B, 0)).

The first problem is that np.linalg.pinv is the pseudo-inverse, A+, not the multiplicative inverse, A^-1. Using a completely different operation just because it doesn't give you an error doesn't solve the problem.

So, how do you calculate the multiplicative inverse? Well, you can't. In general, the multiplicative inverse doesn't exist for non-square matrices, so given a 5x10 A, you're stuck right at the beginning.

Anyway, the second problem comes from the fact that your br isn't a column vector. If you want to think in matrix terms, as the paper does, it's a row vector, 10x1 instead of 1x10. If you want to think in numpy ndarray terms, it's a 1D (10,) array instead of a 2D (1, 10) array. If you think of the operation in matrix multiplication terms, you can't multiply a 10x5 matrix with a 10x1 matrix; if you think of it in NumPy terms as the multidimensional dot product, you can't multiply a (10, 5) array with a (10,) array.

It's true that you can extend the dot product to specifically the domain of MxN matrices vs. M vectors, and under that definition your multiplication would make sense. But that's not the definition used by either the paper's standard matrix multiplication notation or NumPy's dot function. So, what can you do? Well, note that the operation you're trying to do is commutative, so swapping the order of operands is perfectly legal—and if you do that, then it does happen to correspond to the general dot product. So, you could write this as np.dot(np.sum(B, 0), np.linalg.pinv(A)) and get the result you want. And there are a number of other ways you could transform the arrays that are idempotent in your matrix-vs.-vector multiplication domain but meaningful for np.dot, and they will all get you the same result. For example, np.dot(np.linalg.pinv(A).T, np.sum(B, 0)) will also work.

I'm also not sure why you're using dot product in the first place. I don't see anything in the notation to imply that

But all of this is a sideshow; if you inverted A properly, you would have something with the same dimensions as A, and multiply a 5x10 matrix with a 10x1 vector, or a (5, 10) array with a (10,) array, is already perfectly well defined. The only problem is that, again, you can't generally invert non-square matrices, so there's no way you can actually get to this place.

So, the real solution is to go back to wherever you decided on those shapes for A and B and try again.

In particular, it's pretty clear from the illustration in the paper showing the derivation of A and B from the larger matrix that the height of A is the height of B, and the width of A is the width of B.T, which is of course the height of B again.

Also, if the larger matrix is supposed to symmetric, and A is the upper left corner of a symmetric matrix, A has to be symmetric.

I also think you've mixed up row-column order and x-y order a few times, and bc is supposed to the column sums of B, not the column sums of B.T (which would just be the row sums of B, flipped into a row vector instead of a column vector).

While we're at it, let's use methods and operators where possible instead of writing everything in the longest possible way.

So, I think what you wanted is something like this:

A = np.random.random_sample((4, 4)) # square
A = (A + A.T) / 2 # and symmetric
B = np.random.random_sample((4, 10))
ar = A.sum(1)
br = B.sum(1)
bc = B.sum(0) # not B.T.sum(0), that's just br again!
d1 = ar + br
d2 = bc + np.dot(B.T, np.dot(np.linalg.inv(A), br))

Without actually reading the paper I can't be sure this is what you actually want, but this looks like it fits with a quick skim of those two pages, and it runs without any errors, so hopefully you can at least look at the results and see if they are what you want.

Improve this answer
Sign up to request clarification or add additional context in comments.

10 Comments

the thing is that the equations do not mention that i should transpose the inv(A) but i should use it like this. I mean, i could do some tricks to get the job done but in theory it is not like this. :s
@GeorgeLaza: I don't know what equations you're looking at or what form they're in, since you haven't shown them. Maybe they're expecting you to have a row vector rather than a column vector. Or maybe, as I explained in the answer, they're relying on the fact that the dot product of a matrix and a vector is unambiguous (unlike the general N-dimensional case).
i'm trying to do the Nystrom approximation as explained in the corresponding chapter of this thesis[1]. I would be glad if you could help me on this. [1]:people.csail.mit.edu/lauren/publications/…
@GeorgeLaza: First, if that's relevant to the question, edit the question to include it, don't bury it in a comment. Second, do you not understand my explanation of why it probably doesn't matter, or do you just not agree with it? At any rate, I'm not going to read a 200-page thesis to try to figure out where it explains what you're trying to do so I can try to guess what you may be missing in your attempt to translate it in some way you haven't explained.
i read your answer but i think i am making a logical error and not a syntax one, that's why i wanted another pair of eyes to see the logic behind it. The Nystrom method is only 2 pages long that's why i post the link of the thesis (p.90-92). Of course i don't expect you to read the whole thing :)
|
1

You are summing over the first dimension of B, so the shape is 10, the size of the second dimension of B.

You can calculate

np.dot( np.sum(B, 0), np.linalg.pinv(A))

but this gives you a vector with 5 elements, but B_T has only a size of 4. So something doesn't fit in your sample data.

Improve this answer

Comments

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.