I'm converting MATLAB code to Python
This is my code in python:
import numpy as np
import math
n=150
L=1
inter=L/n
y=np.linspace(inter/2,L-inter/2,n).transpose()
E=(210000000000)*np.ones(n)
Rho=7800*np.ones(n)
PI=math.pi
A=np.exp( 5+2*y*(np.sin(2*PI*y/L)) )*0.000001
This works fine up until this point with no difference in values or issues until I have to execute this piece of MATLAB code.
Mass=sum(Rho*inter.*A)
I tried the same using np.sum(Rho*inter*A) and just Rho*inter*A
The first case I got a single answer 1.0626206716847877 but MATLAB returns a 150 element array.
In the scond case I got an ndarray like I wanted but the values were not the same as what I got in MATLAB.
Values I got in MATLAB : matlab values pastebin
Values I got in python : python values pastebin
What am I doing wrong?
(Rho[:,None]*inter*A).sum(axis=0)
matches your MATLAB pastebin.
Or using einsum to sort out the axes:
np.einsum('i,j->j', Rho,inter*A)
which just reduces to:
Rho.sum() * inter*A
Is that really what you are trying to do in MATLAB?
It might help if you showed the actual MATLAB code used to create Rho, A etc.
Mass=sum(Rho*inter.*A)
What's the size of Rho and A in MATLAB? One may be [1x150], but the other? Is Rho [1x150] also, or [150x150]. The * is matrix multiplication, like # in numpy, but .* is elementwise.
In the numpy code y, Rho and A all have shape (150,). The transpose on y does nothing. Rho*inter*A is elementwise multiplication producing a (150,) as well.
NumPy always sums all elements of a matrix. MATLAB's default is column-based, i.e. all of your 150 columns sum to a single total, hence the array. Use sum(matrix,'all'); in MATLAB to sum over all elements in a matrix. If you have a MATLAB older than 2018b, use sum(matrix(:)), i.e. store your matrix in a temporary variable, then flatten it to a column before summing.
To sum over columns in Python, specify the axis, being 0: np.sum(matrix,axis=0)
numpy.sum():
Axis or axes along which a sum is performed. The default, axis=None, will sum all of the elements of the input array.
sum() from MATLAB:
S = sum(A) returns the sum of the elements of A along the first array dimension whose size does not equal 1.
If A is a matrix, then sum(A) returns a row vector containing the sum of each column.
S = sum(A,'all') computes the sum of all elements of A. This syntax is valid for MATLAB® versions R2018b and later.
To prevent this kind of unclarities, I prefer to always specify which direction to sum over, i.e. sum(matrix,1) for MATLAB and np.sum(matrix,axis=0) for NumPy, regardless of the default.
I think that in MATLAB using sum on a matrix you will get the sum of its individual columns and you will end up with an array with its number of elements equal to that of the columns. Use one more sum command in MATLAB: sum(sum(M)), which is the equivalent of np.sum(M) in Python.
Related
I am facing a mystery right now. I get strange results in some program and I think it may be related to the computation since I got different results with my functions compared to manual computation.
This is from my program, I am printing the values pre-computation :
print("\nPrecomputation:\nmatrix\n:", matrix)
tmp = likelihood_left * likelihood_right
print("\nconditional_dep:", tmp)
print("\nfinal result:", matrix # tmp)
I got the following output:
Precomputation:
matrix:
[array([0.08078721, 0.5802404 , 0.16957052, 0.09629893, 0.07310294])
array([0.14633129, 0.45458744, 0.20096238, 0.02142105, 0.17669784])
array([0.41198731, 0.06197812, 0.05934063, 0.23325626, 0.23343768])
array([0.15686545, 0.29516415, 0.20095091, 0.14720275, 0.19981674])
array([0.15965914, 0.18383683, 0.10606946, 0.14234812, 0.40808645])]
conditional_dep: [0.01391123 0.01388155 0.17221067 0.02675524 0.01033257]
final result: [0.07995043 0.03485223 0.02184015 0.04721548 0.05323298]
The thing is when I compute the following code:
matrix = [np.array([0.08078721, 0.5802404 , 0.16957052, 0.09629893, 0.07310294]),
np.array([0.14633129, 0.45458744, 0.20096238, 0.02142105, 0.17669784]),
np.array([0.41198731, 0.06197812, 0.05934063, 0.23325626, 0.23343768]),
np.array([0.15686545, 0.29516415, 0.20095091, 0.14720275, 0.19981674]),
np.array([0.15965914, 0.18383683, 0.10606946, 0.14234812, 0.40808645])]
tmp = np.asarray([0.01391123, 0.01388155, 0.17221067, 0.02675524, 0.01033257])
matrix # tmp
The values in use are exactly the same as they should be in the computation before but I get the following result:
array([0.04171218, 0.04535276, 0.02546353, 0.04688848, 0.03106443])
This result is then obviously different than the previous one and is the true one (I computed the dot product by hand).
I have been facing this problem the whole day and I did not find anything useful online. If any of you have any even tiny idea where it can come from I'd be really happy :D
Thank's in advance
Yann
PS: I can show more of the code if needed.
PS2: I don't know if it is relevant but this is used in a dynamic programming algorithm.
To recap our discussion in the comments, in the first part ("pre-computation"), the following is true about the matrix object:
>>> matrix.shape
(5,)
>>> matrix.dtype
dtype('O') # aka object
And as you say, this is due to matrix being a slice of a larger, non-uniform array. Let's recreate this situation:
>>> matrix = np.array([[], np.array([0.08078721, 0.5802404 , 0.16957052, 0.09629893, 0.07310294]), np.array([0.14633129, 0.45458744, 0.20096238, 0.02142105, 0.17669784]), np.array([0.41198731, 0.06197812, 0.05934063, 0.23325626, 0.23343768]), np.array([0.15686545, 0.29516415, 0.20095091, 0.14720275, 0.19981674]), np.array([0.15965914, 0.18383683, 0.10606946, 0.14234812, 0.40808645])])[1:]
It is now not a matrix with scalars in rows and columns, but a column vector of column vectors. Technically, matrix # tmp is an operation between two 1-D arrays and hence NumPy should, according to the documentation, calculate the inner product of the two. This is true in this case, with the convention that the sum be over the first axis:
>>> np.array([matrix[i] * tmp[i] for i in range(5)]).sum(axis=0)
array([0.07995043, 0.03485222, 0.02184015, 0.04721548, 0.05323298])
>>> matrix # tmp
array([0.07995043, 0.03485222, 0.02184015, 0.04721548, 0.05323298])
This is essentially the same as taking the transpose of the proper 2-D matrix before the multiplication:
>>> np.stack(matrix).T # tmp
array([0.07995043, 0.03485222, 0.02184015, 0.04721548, 0.05323298])
Equivalently, as noted by #jirasssimok:
>>> tmp # np.stack(matrix)
array([0.07995043, 0.03485222, 0.02184015, 0.04721548, 0.05323298])
Hence the erroneous or unexpected result.
As you have already resolved to do in the comments, this can be avoided in the future by ensuring all matrices are proper 2-D arrays.
It looks like you got the operands switched in one of your matrix multiplications.
Using the same values of matrix and tmp that you provided, matrix # tmp and tmp # matrix provide the two results you showed.1
matrix = [np.array([0.08078721, 0.5802404 , 0.16957052, 0.09629893, 0.07310294]),
np.array([0.14633129, 0.45458744, 0.20096238, 0.02142105, 0.17669784]),
np.array([0.41198731, 0.06197812, 0.05934063, 0.23325626, 0.23343768]),
np.array([0.15686545, 0.29516415, 0.20095091, 0.14720275, 0.19981674]),
np.array([0.15965914, 0.18383683, 0.10606946, 0.14234812, 0.40808645])]
tmp = np.asarray([0.01391123, 0.01388155, 0.17221067, 0.02675524, 0.01033257])
print(matrix # tmp) # [0.04171218 0.04535276 0.02546353 0.04688848 0.03106443]
print(tmp # matrix) # [0.07995043 0.03485222 0.02184015 0.04721548 0.05323298]
To make it a little more obvious what your code is doing, you might also consider using np.dot instead of #. If you pass matrix as the first argument and tmp as the second, it will have the result you want, and make it more clear that you're conceptually calculating dot products rather than multiplying matrices.
As an additional note, if you're performing matrix operations on matrix, it might be better if it was a single two-dimensional array instead of a list of 1-dimensional arrays. this will prevent errors of the sort you'll see right now if you try to run matrix # matrix. This would also let you say matrix.dot(tmp) instead of np.dot(matrix, tmp) if you wanted to.
(I'd guess that you can use np.stack or a similar function to create matrix, or you can call np.stack on matrix after creating it.)
1 Because tmp has only one dimension and matrix has two, NumPy can and will treat tmp as whichever type of vector makes the multiplication work (using broadcasting). So tmp is treated as a column vector in matrix # tmp and a row vector in tmp # matrix.
I am trying to multiply all the row values and column values of a 2 dimensional numpy array with an explicit for-loop:
product_0 = 1
product_1 = 1
for x in arr:
product_0 *= x[0]
product_1 *= x[1]
I realize the product will blow up to become an extremely large number but from my previous experience python has had no memory problem dealing very very extremely large numbers.
So from what I can tell this is a problem with numpy except I am not storing the gigantic product in a numpy array or any numpy data type for that matter its just a normal python variable.
Any idea how to fix this?
Using non inplace multiplication hasn't helped product_0 = x[0]*product_0
Python int are represented in arbitrary precision, so they cannot overflow. But numpy uses C++ under the hood, so the highest long signed integer is with fixed precision, 2^63 - 1. Your number is far beyond this value, having in average ((716-1)/2)^86507).
When you, in the for loop, extract the x[0] this is still a numpy object. To use the full power of python integers you need to clearly assign it as python int, like this:
product_0 = 1
product_1 = 1
for x in arr:
t = int(x[0])
product_0 = product_0 * t
and it will not overflow.
Following your comment, which makes your question more specific, your original problem is to calculate the geometric mean of the array for each row/column. Here the solution:
I generate first an array that has the same properties of your array:
arr = np.resize(np.random.randint(1,716,86507*2 ),(86507,2))
Then, calculate the geometric mean for each column/row:
from scipy import stats
gm_0 = stats.mstats.gmean(arr, axis = 0)
gm_1 = stats.mstats.gmean(arr, axis = 1)
gm_0 will be an array that contains the geometric mean of the xand y coordinates. gm_1 instead contains the geometric mean of the rows.
Hope this solves your problem!
You say
So from what I can tell this is a problem with numpy except I am not storing the gigantic product in a numpy array or any numpy data type for that matter its just a normal python variable.
Your product may not be a NumPy array, but it is using a NumPy data type. x[0] and x[1] are NumPy scalars, and multiplying a Python int by a NumPy scalar produces a NumPy scalar. NumPy integers have a finite range.
While you technically could call int on x[0] and x[1] to get a Python int, it'd probably be better to avoid needing such huge ints. You say you're trying to perform this multiplication to compute a geometric mean; in that case, it'd be better to compute the geometric mean by transforming to and from logarithms, or to use scipy.stats.mstats.gmean, which uses logarithms under the hood.
Numpy is compiled for 32 bit and not 64 bit , so while Python can handle this numpy will overflow for smaller values , if u want to use numpy then I recommend that you build it from source .
Edit
After some testing with
import numpy as np
x=np.abs(np.random.randn(1000,2)*1000)
np.max(x)
prod1=np.dtype('int32').type(1)
prod2=np.dtype('int32').type(1)
k=0
for i,j in x:
prod1*=i
prod2*=j
k+=1
print(k," ",prod1,prod2)
1.797693134e308 is the max value (to this many digits my numpy scalar was able to take)
if you run this you will see that numpy is able to handle quite a large value but when you said your max value is around 700 , even with a 1000 values my scalar overflowed.
As for how to fix this , rather than doing this manually the answer using scipy seems more viable now and is able to get the answer so i suggest that you go forward with that
from scipy.stats.mstats import gmean
x=np.abs(np.random.randn(1000,2)*1000)
print(gmean(x,axis=0))
You can achieve what you want with the following command in numpy:
import numpy as np
product_0 = np.prod(arr.astype(np.float64))
It can still reach np.inf if your numbers are large enough, but that can happen for any type.
I tried to solve a PDE numerically and in the course of this I faced the problem of a triple-nested for loop resembling the 3 spatial dimension. This construct is nested in another time loop, so you can imagine that the computing takes forever for sufficient large node numbers. The code block looks like this
for jy in range(0,cy-1):
for jx in range(0,cx-1):
for jz in range(0,cz-1):
T[n+1,jx,jy,jz] = T[n,jx,jy,jz] + s*(T[n,jx-1,jy,jz] - 2*T[n,jx,jy,jz] + T[n,jx+1,jy,jz]) + s*(T[n,jx,jy-1,jz] - 2*T[n,jx,jy,jz] + T[n,jx,jy+1,jz]) + s*(T[n,jx,jy,jz-1] - 2*T[n,jx,jy,jz] + T[n,jx,jy,jz+1])
It might look intimidating at first, but is quite easy. I have a 3 dimensional matrix representing a solid bulk material, where each point represents the current temperature. The iteratively calculated next temperature at each point is calculated taking into account each point next to that point - so 6 in total. In the case of a 1-dimensional solid the solution is just a simple matrix multiplication. Is there any chance to represent the 3-loop-system above in a simple matrix solution like in the 1D case?
Best regards!
With numpy you can easily do these kinds of matrix operations,
e.g for a 3x3 matrix
import numpy as np
T = np.random.random((3,3,3))
T = T*T - 2*T ... etc.
First off, you need to be a bit more careful with your terminology. A "matrix" is a 2-Dimensional array of numbers. So you are really talking about an array. Numpy, or better yet Scipy, has an data type called an ndarray. You need to be very careful manipulating them, because although they are sometimes used to represent matrices, there are operations that can be performed on 2-D arrays that are not mathematically legal for matrices.
I strongly recommend you use # and not * to perform multiplication of 1- or 2-D matrices, and be sure to add code to check that the operations you are doing are legal mathematically. As a trivial example, Python lets you add a 1 x n or an n x 1 vector to an n x n matrix, even though that is not mathematically correct. The reason it allows it is, as intimated above, because there is no true matrix type in Python.
It very well may be that you can reformulate your problem to use a 3-D array, and by experimentation find the particular operation you are trying to perform. Just keep in mind that the rules of linear algebra are only casually applied in Python.
I have a problem. I have two arrays with the size of 82248x20 and if i do the following in Matlab
A=X'*Y
it will give me 6.152847328855238e-18 for the second value.
If i do it in Python with anything like
test=scipy.io.loadmat('wohin.mat')
X=test['X']
Y=test['Y']
A=np.transpose(X)#Y
A=np.dot(np.transpose(X),Y)
A=np.matmul(np.transpose(X),Y)
i get the value 1.9233746539892849e-16 for the second value and if i do the calculation with
for i in range(0,82248):
t=t+np.transpose(Y)[0,i]*X[i,1]
i get 3.3664996263355106e-15 for the second value of row one. So where is my misunderstanding or the difference between the three methods. The last one has some rounding errors perhaps, but the two other ones should give me the same result or not?
Mat file with the matrices is here
The two matrices X and Y are identical matrices with columns forming what appears to be an orthonormal basis. Therefore, you should expect transpose(X)*Y to be an identity matrix. All the off-diagonal elements should be zero and only differ from zero due to rounding errors.
That said, the differences you observe simply imply that the various implementations of matrix multiplication differ from each other in some ways. For example, the order in which sums are taken can affect the final result.
Example (MATLAB):
>> sum(X(:,1).*Y(:,2))
ans =
3.366499626335511e-15
>> sum(flipud(X(:,1)).*flipud(Y(:,2)))
ans =
3.366880519846534e-15
In this example we manually take the inner product of two large orthogonal vectors. By flipping the vectors the result shouldn't change, however, due to rounding errors we get slightly different results.
I need to invert a large, dense matrix which I hoped to use Scipy's gmres to do. Fortunately, the dense matrix A follows a pattern and I do not need to store the matrix in memory. The LinearOperator class allows us to construct an object which acts as the matrix for GMRES and can compute directly the matrix vector product A*v. That is, we write a function mv(v) which takes as input a vector v and returns mv(v) = A*v. Then, we can use the LinearOperator class to create A_LinOp = LinearOperator(shape = shape, matvec = mv). We can put the linear operator into the Scipy gmres command to evaluate the matrix vector products without ever having to fully load A into memory.
The documentation for the LinearOperator is found here: LinearOperator Documentation.
Here is my problem: to write the routine to compute the matrix vector product mv(v) = A*v, I need another input vector C. The entries in A are of the form A[i,j] = f(C[i] - C[j]). So, what I really want is for mv to be of two inputs, one fixed vector input C, and one variable input v which we want to compute A*v.
MATLAB has a similar setup, where would write x = gmres(#(v) mv(v,C),b) where b is the right hand side of the problem Ax = b, , and mv is the function that takes as variable input v which we want to compute A*v and C is the fixed, known vector which we need for the assembly of A.
My problem is that I can't figure out how to allow the LinearOperator class to accept two inputs, one variable and one "fixed" like I can in MATLAB.
Is there a way to do the analogous operation in SciPy? Alternatively, if anyone knows of a better way of inverting a large, dense matrix (50000, 50000) where the entries follow a pattern, I would greatly appreciate any suggestions.
Thanks!
EDIT: I should have stated this information actually. The matrix is actually (in block form) [A C; C^T 0], where A is N x N (N large) and C is N x 3, and the 0 is 3 x 3 and C^T is the transpose of C. This array C is the same array as the one mentioned above. The entries of A follow a pattern A[i,j] = f(C[i] - C[j]).
I wrote mv(v,C) to go row by row construct A*v[i] for i=0,N, by computing sum f(C[i]-C[j)*v[j] (actually, I do numpy.dot(FC,v) where FC[j] = f(C[i]-C[j]) which works well). Then, at the end doing the computations for the C^T rows. I was hoping to eventually replace the large for loop with a multiprocessing call to parallelize the for loop, but that's a future thing to consider. I will also look into using Cython to speed up the computations.
This is very late, but if you're still interested...
Your A matrix must be very low rank since it's a nonlinearly transformed version of a rank-2 matrix. Plus it's symmetric. That means it's trivial to inverse: get the truncated eigenvalue decompostion with, say, 5 eigenvalues: A = U*S*U', then invert that: A^-1 = U*S^-1*U'. S is diagonal so this is inexpensive. You can get the truncated eigenvalue decomposition with eigh.
That takes care of A. Then for the rest: use the block matrix inversion formula. Looks nasty, but I will bet you 100,000,000 prussian francs that it's 50x faster than the direct method you were using.
I faced the same situation (some years later than you) of trying to use more than one argument to LinearOperator, but for another problem. The solution I found was the use of global variables, to avoid passing the variables as arguments to the function.