firstly, apologize for little cryptic title to my question. Let me try to explain my need:-
I am reading two features namely X1, X2 from a CSV file. I have a training set of data in a csv file containing 1000 records with each line corresponding to the value of X1, X2. To make my training set fit better to my machine learning code, I want to do feature mapping that would take X1, X2 and create polynomial terms to the power of 4. for example if X1 =a, X2=b, I want to add newer features a^2, a*b, b^2, a^3,a^2*b,a*b^2,a^4...and so on.
Now if I read them as a numpy matrix , I want to see the data like this:
[ [ 1 a b a^2 a*b, b^2 a^3 a^2*b......]
[.... ............ ............ ]
[ ..
..] ]
Note that the number of rows are fixed , but the number of columns are determined by the degree selected. Also first three columns need to be
[[1 a b ..]
[1 c d ..]
..
..]
The pseudo code I am thinking of is as follows:-
def poly(X): # where X is a numpy matrix with X1, X2 columns,
degree = 4;
r= X.shape[0]
c=1 # number of columns
val_matrix= np.ones(shape=(r,c)) # creating a (r,1) matrix init with 1s
# *start of psuedo code*
while i<=degree:
while j <=i:
val_matrix[:, c+1] = (X1.^(i-j)).*(X2.^j)
I am not sure how to get this working in python?. would appreciate some suggestion. Note that ^ refers to the power of.
Starting with two vectors X1 and X2 you could create the monomials:
X1p = X1[:, None]**np.arange(max_deg + 1)
X2p = X2[:, None]**np.arange(max_deg + 1)
and then combine them using mgrid
i, j = np.mgrid[:max_deg + 1,:max_deg + 1]
m = i+j <= max_deg
result = X1p[:, i[m]]*X2p[:, j[m]]
Alternatively you could apply the indices directly to X1 and X2:
result = X1[:, None]**i[m] * X2[:, None]**j[m]
This requires fewer lines of code but uses more multiplications.
If the number of multiplications is a concern, X1p and X2p could also be computed cheaper; X1p:
X1p = np.empty((len(X1), max_deg + 1), X1.dtype)
X1p[:, 0] = 1
X1p[:, 1:] = X1[:, None]
np.multiply.accumulate(X1p[:,1:], axis=-1, out=X1p[:, 1:])
and similar for X2p
Related
I'm creating a matrix of order [NBxNB] where each element is another matrix of order [3x3], and all the elements inside these matrices are complex numbers.
These [3x3] matrices are built with some FOR loops and must be positioned on specific positions on the [NBxNB] matrix. After all the [3x3] are correct positioned,
I need to transform the blocks on a single [3NBx3NB] matrix, evaluate the inverse matrix and divide the inverse [3NBx3NB] matrix into the same [NBxNB] structure with the [3x3] submatrices.
I need this division because I want to be able to select the submatrices individualy to perform a few calculations after the inversion.
The main procedure to build the matrix and inverse is shown below, with some zero data just to exemplify the code.
#This is the function to divide the inverse in blocks - Got here on StackOverflow but don't remeber from which post...
def blockshaped(arr, nrows, ncols):
h, w = arr.shape
assert h % nrows == 0, "{} rows is not evenly divisble by {}".format(h, nrows)
assert w % ncols == 0, "{} cols is not evenly divisble by {}".format(w, ncols)
return (arr.reshape(h//nrows, nrows, -1, ncols)
.swapaxes(1,2)
.reshape(-1, nrows, ncols))
Y = np.zeros([NB,NB], dtype=object)
for i in range(NB):
for j in range(NB):
Y[i][j] = np.zeros([3,3])
Y[j][i] = np.zeros([3,3])
AUX = np.zeros(NB, dtype=object)
for i in range(NB):
AUX[i] = np.vstack(Y[:,i]).astype(complex)
T = np.hstack(AUX).astype(complex)
X = np.linalg.inv(T)
Z = blockshaped(X,3,3)
This code is working fine but I was figuring if there is a more 'pythonic way' of doing so.
I have a code that works perfectly well but I wish to speed up the time it takes to converge. A snippet of the code is shown below:
def myfunction(x, i):
y = x + (min(0, target[i] - data[i, :]x))*data[i]/(norm(data[i])**2))
return y
rows, columns = data.shape
start = time.time()
iterate = 0
iterate_count = []
norm_count = []
res = 5
x_not = np.ones(columns)
norm_count.append(norm(x_not))
iterate_count.append(0)
while res > 1e-8:
for row in range(rows):
y = myfunction(x_not, row)
x_not = y
iterate += 1
iterate_count.append(iterate)
norm_count.append(norm(x_not))
res = abs(norm_count[-1] - norm_count[-2])
print('Converge at {} iterations'.format(iterate))
print('Duration: {:.4f} seconds'.format(time.time() - start))
I am relatively new in Python. I will appreciate any hint/assistance.
Ax=b is the problem we wish to solve. Here, 'A' is the 'data' and 'b' is the 'target'
Ugh! After spending a while on this I don't think it can be done the way you've set up your problem. In each iteration over the row, you modify x_not and then pass the updated result to get the solution for the next row. This kind of setup can't be vectorized easily. You can learn the thought process of vectorization from the failed attempt, so I'm including it in the answer. I'm also including a different iterative method to solve linear systems of equations. I've included a vectorized version -- where the solution is updated using matrix multiplication and vector addition, and a loopy version -- where the solution is updated using a for loop to demonstrate what you can expect to gain.
1. The failed attempt
Let's take a look at what you're doing here.
def myfunction(x, i):
y = x + (min(0, target[i] - data[i, :] # x)) * (data[i] / (norm(data[i])**2))
return y
You subtract
the dot product of (the ith row of data and x_not)
from the ith row of target,
limited at zero.
You multiply this result with the ith row of data divided my the norm of that row squared. Let's call this part2
Then you add this to the ith element of x_not
Now let's look at the shapes of the matrices.
data is (M, N).
target is (M, ).
x_not is (N, )
Instead of doing these operations rowwise, you can operate on the entire matrix!
1.1. Simplifying the dot product.
Instead of doing data[i, :] # x, you can do data # x_not and this gives an array with the ith element giving the dot product of the ith row with x_not. So now we have data # x_not with shape (M, )
Then, you can subtract this from the entire target array, so target - (data # x_not) has shape (M, ).
So far, we have
part1 = target - (data # x_not)
Next, if anything is greater than zero, set it to zero.
part1[part1 > 0] = 0
1.2. Finding rowwise norms.
Finally, you want to multiply this by the row of data, and divide by the square of the L2-norm of that row. To get the norm of each row of a matrix, you do
rownorms = np.linalg.norm(data, axis=1)
This is a (M, ) array, so we need to convert it to a (M, 1) array so we can divide each row. rownorms[:, None] does this. Then divide data by this.
part2 = data / (rownorms[:, None]**2)
1.3. Add to x_not
Finally, we're adding each row of part1 * part2 to the original x_not and returning the result
result = x_not + (part1 * part2).sum(axis=0)
Here's where we get stuck. In your approach, each call to myfunction() gives a value of part1 that depends on target[i], which was changed in the last call to myfunction().
2. Why vectorize?
Using numpy's inbuilt methods instead of looping allows it to offload the calculation to its C backend, so it runs faster. If your numpy is linked to a BLAS backend, you can extract even more speed by using your processor's SIMD registers
The conjugate gradient method is a simple iterative method to solve certain systems of equations. There are other more complex algorithms that can solve general systems well, but this should do for the purposes of our demo. Again, the purpose is not to have an iterative algorithm that will perfectly solve any linear system of equations, but to show what kind of speedup you can expect if you vectorize your code.
Given your system
data # x_not = target
Let's define some variables:
A = data.T # data
b = data.T # target
And we'll solve the system A # x = b
x = np.zeros((columns,)) # Initial guess. Can be anything
resid = b - A # x
p = resid
while (np.abs(resid) > tolerance).any():
Ap = A # p
alpha = (resid.T # resid) / (p.T # Ap)
x = x + alpha * p
resid_new = resid - alpha * Ap
beta = (resid_new.T # resid_new) / (resid.T # resid)
p = resid_new + beta * p
resid = resid_new + 0
To contrast the fully vectorized approach with one that uses iterations to update the rows of x and resid_new, let's define another implementation of the CG solver that does this.
def solve_loopy(data, target, itermax = 100, tolerance = 1e-8):
A = data.T # data
b = data.T # target
rows, columns = data.shape
x = np.zeros((columns,)) # Initial guess. Can be anything
resid = b - A # x
resid_new = b - A # x
p = resid
niter = 0
while (np.abs(resid) > tolerance).any() and niter < itermax:
Ap = A # p
alpha = (resid.T # resid) / (p.T # Ap)
for i in range(len(x)):
x[i] = x[i] + alpha * p[i]
resid_new[i] = resid[i] - alpha * Ap[i]
# resid_new = resid - alpha * A # p
beta = (resid_new.T # resid_new) / (resid.T # resid)
p = resid_new + beta * p
resid = resid_new + 0
niter += 1
return x
And our original vector method:
def solve_vect(data, target, itermax = 100, tolerance = 1e-8):
A = data.T # data
b = data.T # target
rows, columns = data.shape
x = np.zeros((columns,)) # Initial guess. Can be anything
resid = b - A # x
resid_new = b - A # x
p = resid
niter = 0
while (np.abs(resid) > tolerance).any() and niter < itermax:
Ap = A # p
alpha = (resid.T # resid) / (p.T # Ap)
x = x + alpha * p
resid_new = resid - alpha * Ap
beta = (resid_new.T # resid_new) / (resid.T # resid)
p = resid_new + beta * p
resid = resid_new + 0
niter += 1
return x
Let's solve a simple system to see if this works first:
2x1 + x2 = -5
−x1 + x2 = -2
should give a solution of [-1, -3]
data = np.array([[ 2, 1],
[-1, 1]])
target = np.array([-5, -2])
print(solve_loopy(data, target))
print(solve_vect(data, target))
Both give the correct solution [-1, -3], yay! Now on to bigger things:
data = np.random.random((100, 100))
target = np.random.random((100, ))
Let's ensure the solution is still correct:
sol1 = solve_loopy(data, target)
np.allclose(data # sol1, target)
# Output: False
sol2 = solve_vect(data, target)
np.allclose(data # sol2, target)
# Output: False
Hmm, looks like the CG method doesn't work for badly conditioned random matrices we created. Well, at least both give the same result.
np.allclose(sol1, sol2)
# Output: True
But let's not get discouraged! We don't really care if it works perfectly, the point of this is to demonstrate how amazing vectorization is. So let's time this:
import timeit
timeit.timeit('solve_loopy(data, target)', number=10, setup='from __main__ import solve_loopy, data, target')
# Output: 0.25586539999994784
timeit.timeit('solve_vect(data, target)', number=10, setup='from __main__ import solve_vect, data, target')
# Output: 0.12008900000000722
Nice! A ~2x speedup simply by avoiding a loop while updating our solution!
For larger systems, this will be even better.
for N in [10, 50, 100, 500, 1000]:
data = np.random.random((N, N))
target = np.random.random((N, ))
t_loopy = timeit.timeit('solve_loopy(data, target)', number=10, setup='from __main__ import solve_loopy, data, target')
t_vect = timeit.timeit('solve_vect(data, target)', number=10, setup='from __main__ import solve_vect, data, target')
print(N, t_loopy, t_vect, t_loopy/t_vect)
This gives us:
N t_loopy t_vect speedup
00010 0.002823 0.002099 1.345390
00050 0.051209 0.014486 3.535048
00100 0.260348 0.114601 2.271773
00500 0.980453 0.240151 4.082644
01000 1.769959 0.508197 3.482822
Though code runs well, but results not displayed. Did I miss anything?
# Independent variables
x1 = 'a'
x2 = 'b'
# Dependent variable
y = 'c'
# Pairings
x1y = cor.loc[ x1, y ]
x2y = cor.loc[ x2, y ]
x1x2 = cor.loc[ x1, x2 ]
Rx1x2y = math.sqrt((abs(x1y**2) + abs(x2y**2) - 2*x1y*x2y*x1x2) / (1-
abs(x1x2**2)) )
R2 = Rx1x2y**2
# Calculate adjusted R-squared
n = len(data) # Number of rows
k = 2 # Number of independent variables
R2_adj = 1 - ( ((1-R2)*(n-1)) / (n-k-1) )
Above codes ran well nut no results displayed here. I was expecting R2,R2_adj values in output
Say, I have a binary (adjacency) matrix A of dimensions nxn and another matrix U of dimensions nxl. I use the following piece of code to compute a new matrix that I need.
import numpy as np
from numpy import linalg as LA
new_U = np.zeros_like(U)
for idx, a in np.ndenumerate(A):
diff = U[idx[0], :] - U[idx[1], :]
if a == 1.0:
new_U[idx[0], :] += 2 * diff
elif a == 0.0:
norm_diff = LA.norm(U[idx[0], :] - U[idx[1], :])
new_U[idx[0], :] += -2 * diff * np.exp(-norm_diff**2)
return new_U
This takes quite a lot of time to run even when n and l are small. Is there a better way to rewrite (vectorize) this code to reduce the runtime?
Edit 1: Sample input and output.
A = np.array([[0,1,0], [1,0,1], [0,1,0]], dtype='float64')
U = np.array([[2,3], [4,5], [6,7]], dtype='float64')
new_U = np.array([[-4.,-4.], [0,0],[4,4]], dtype='float64')
Edit 2: In mathematical notation, I am trying to compute the following:
where u_ik = U[i, k],u_jk = U[j, k], and u_i = U[i, :]. Also, (i,j) \in E corresponds to a == 1.0 in the code.
Leveraging broadcasting and np.einsum for the sum-reductions -
# Get pair-wise differences between rows for all rows in a vectorized manner
Ud = U[:,None,:]-U
# Compute norm L1 values with those differences
L = LA.norm(Ud,axis=2)
# Compute 2 * diff values for all rows and mask it with ==0 condition
# and sum along axis=1 to simulate the accumulating behaviour
p1 = np.einsum('ijk,ij->ik',2*Ud,A==1.0)
# Similarly, compute for ==1 condition and finally sum those two parts
p2 = np.einsum('ijk,ij,ij->ik',-2*Ud,np.exp(-L**2),A==0.0)
out = p1+p2
Alternatively, use einsum for computing squared-norm values and using those to get p2 -
Lsq = np.einsum('ijk,ijk->ij',Ud,Ud)
p2 = np.einsum('ijk,ij,ij->ik',-2*Ud,np.exp(-Lsq),A==0.0)
I have two functions that compute the same metric. One ends up using a list comprehension to cycle through a calculation, the other uses only numpy tensor operations. The functions take in a (N, 3) array, where N is the number of points in 3D space. When N <~ 3000 the tensor function is faster, when N >~ 3000 the list comprehension is faster. Both seem to have linear time complexity in terms of N i.e two time-N lines cross at N=~3000.
def approximate_area_loop(section, num_area_divisions):
n_a_d = num_area_divisions
interp_vectors = get_section_interp_(section)
a1 = section[:-1]
b1 = section[1:]
a2 = interp_vectors[:-1]
b2 = interp_vectors[1:]
c = lambda u: (1 - u) * a1 + u * a2
d = lambda u: (1 - u) * b1 + u * b2
x = lambda u, v: (1 - v) * c(u) + v * d(u)
area = np.sum([np.linalg.norm(np.cross((x((i + 1)/n_a_d, j/n_a_d) - x(i/n_a_d, j/n_a_d)),\
(x(i/n_a_d, (j +1)/n_a_d) - x(i/n_a_d, j/n_a_d))), axis = 1)\
for i in range(n_a_d) for j in range(n_a_d)])
Dt = section[-1, 0] - section[0, 0]
return area, Dt
def approximate_area_tensor(section, num_area_divisions):
divisors = np.linspace(0, 1, num_area_divisions + 1)
interp_vectors = get_section_interp_(section)
a1 = section[:-1]
b1 = section[1:]
a2 = interp_vectors[:-1]
b2 = interp_vectors[1:]
c = np.multiply.outer(a1, (1 - divisors)) + np.multiply.outer(a2, divisors) # c_areas_vecs_divs
d = np.multiply.outer(b1, (1 - divisors)) + np.multiply.outer(b2, divisors) # d_areas_vecs_divs
x = np.multiply.outer(c, (1 - divisors)) + np.multiply.outer(d, divisors) # x_areas_vecs_Divs_divs
u = x[:, :, 1:, :-1] - x[:, :, :-1, :-1] # u_areas_vecs_Divs_divs
v = x[:, :, :-1, 1:] - x[:, :, :-1, :-1] # v_areas_vecs_Divs_divs
sub_area_norm_vecs = np.cross(u, v, axis = 1) # areas_crosses_Divs_divs
sub_areas = np.linalg.norm(sub_area_norm_vecs, axis = 1) # areas_Divs_divs (values are now sub areas)
area = np.sum(sub_areas)
Dt = section[-1, 0] - section[0, 0]
return area, Dt
Why does the list comprehension version work faster at large N? Surely the tensor version should be faster? I'm wondering if it's something to do with the size of the calculations meaning it's too big to be done in cache? Please ask if I haven't included enough information, I'd really like to get to the bottom of this.
The bottleneck in the fully vectorized function was indeed in the np.linalg.norm as #hpauljs comment suggested.
Norm was used only to get the magnitude of all the vectors contained in axis 1. A much simpler and faster method was to just:
sub_areas = np.sqrt((sub_area_norm_vecs*sub_area_norm_vecs).sum(axis = 1))
This gives exactly the same results and sped up the code by up to 25 times faster than the loop implementation (even when the loop doesn't use linalg.norm either).