If I have an ndarray A of dimensions (2,3,4,5), and an arrays B and C of dimension 3, is there a way I can concisely multiply A with B and add C without a loop, such that:
for idx in range(len(B)):
A[:,idx,:,:] = B[idx] * A[:,idx,:,:] + C[idx]
I imagine there should be a 1 line way of doing it that yields better performance. Is that true?
You can reshape your arrays B and C to have shape (3, 1, 1). The additional reshaping to the shape of A will be performed by Numpy in form of broadcasting:
A = A * B[:, None, None] + C[:, None, None]
Related
If I have a MxN numpy array denoted arr, I wish to index over all elements and adjust the values like so
for m in range(arr.shape[0]):
for n in range(arr.shape[1]):
arr[m, n] += x**2 * np.cos(m) * np.sin(n)
Where x is a random float.
Is there a way to broadcast this over the entire array without needing to loop? Thus, speeding up the run time.
You are just adding zeros, because sin(2*pi*k) = 0 for integer k.
However, if you want to vectorize this, the function np.meshgrid could help you.
Check the following example, where I removed the 2 pi in the trigonometric functions to add something unequal zero.
x = 2
arr = np.arange(12, dtype=float).reshape(4, 3)
n, m = np.meshgrid(np.arange(arr.shape[1]), np.arange(arr.shape[0]), sparse=True)
arr += x**2 * np.cos(m) * np.sin(n)
arr
Edit: use the sparse argument to reduce memory consumption.
You can use nested generators of two-dimensional arrays:
import numpy as np
from random import random
x = random()
n, m = 10,20
arr = [[x**2 * np.cos(2*np.pi*j) * np.sin(2*np.pi*i) for j in range(m)] for i in range(n)]
In [156]: arr = np.ones((2, 3))
Replace the range with arange:
In [157]: m, n = np.arange(arr.shape[0]), np.arange(arr.shape[1])
And change the first array to (2,1) shape. A (2,1) array broadcasts with a (3,) to produce a (2,3) result.
In [158]: A = 0.23**2 * np.cos(m[:, None]) * np.sin(n)
In [159]: A
Out[159]:
array([[0. , 0.04451382, 0.04810183],
[0. , 0.02405092, 0.02598953]])
In [160]: arr + A
Out[160]:
array([[1. , 1.04451382, 1.04810183],
[1. , 1.02405092, 1.02598953]])
The meshgrid suggested in the accepted answer does the same thing:
In [161]: np.meshgrid(m, n, sparse=True, indexing="ij")
Out[161]:
[array([[0],
[1]]),
array([[0, 1, 2]])]
This broadcasting may be clearer with:
In [162]: m, n
Out[162]: (array([0, 1]), array([0, 1, 2]))
In [163]: m[:, None] * 10 + n
Out[163]:
array([[ 0, 1, 2],
[10, 11, 12]])
I used to perform an outer subtraction on two one-dimensional arrays as follows to receive a single two-dimensional arrays that contains all pairs of subtractions:
import numpy as np
a = np.arange(5)
b = np.arange(3)
result = np.subtract.outer(a, b)
assert result.shape == (5, 3)
assert np.all(result == np.array([[aa - bb for bb in b] for aa in a ])) # no rounding errors
Now the state space switches to two dimensions, and I would like to perform the same operation, but only perform each subtraction on the two values on the last axis of the arrays A and B:
import numpy as np
A = np.arange(5 * 2).reshape(-1, 2)
B = np.arange(3 * 2).reshape(-1, 2)
result = np.subtract.outer(A, B)
# Obviously the following does not hold, because here we have got all subtractions, therefore the shape (5, 2, 3, 2)
# I would like to exchange np.subtract.outer such that the following holds:
# assert result.shape == (5, 3, 2)
expected_result = np.array([[aa - bb for bb in B] for aa in A ])
assert expected_result.shape == (5, 3, 2)
# That's what I want to hold:
# assert np.all(result == expected_result) # no rounding errors
Is there a "numpy-only" solution to perform this operation?
You can expand/reshape A to (5, 1, 2) and B to (1, 3, 2) and let the broadcasting do the job:
A[:, None, :] - B[None, :, :]
A[:, None] - B[None, :] does it.
A = np.arange(5 * 2).reshape(-1, 2)
B = np.arange(3 * 2).reshape(-1, 2)
expected_result = np.array([[aa - bb for bb in B] for aa in A ])
C = A[:, None] - B[None, :]
np.allclose(expected_result, C)
#> True
The exact same syntax works for your first example too. This is because with your requirement, you are combining every first axis element of A with every first axis element of B.
I wish to compute the dot product between two 3D tensors along the first dimension. I tried the following einsum notation:
import numpy as np
a = np.random.randn(30).reshape(3, 5, 2)
b = np.random.randn(30).reshape(3, 2, 5)
# Expecting shape: (3, 5, 5)
np.einsum("ijk,ikj->ijj", a, b)
Sadly it returns this error:
ValueError: einstein sum subscripts string includes output subscript 'j' multiple times
I went with Einstein sum after I failed at it with np.tensordot. Ideas and follow up questions are highly welcome!
Your two dimensions of size 5 and 5 do not correspond to the same axes. As such you need to use two different subscripts to designate them. For example, you can do:
>>> res = np.einsum('ijk,ilm->ijm', a, b)
>>> res.shape
(3, 5, 5)
Notice you are also required to change the subscript for axes of size 2 and 2. This is because you are computing the batched outer product (i.e. we iterate on two axes at the same time), not a dot product (i.e. we iterate simultaneously on the two axes).
Outer product:
>>> np.einsum('ijk,ilm->ijm', a, b)
Dot product over subscript k, which is axis=2 of a and axis=1 of b:
>>> np.einsum('ijk,ikm->ijm', a, b)
which is equivalent to a#b.
dot product ... along the first dimension is a bit unclear. Is the first dimension a 'batch' dimension, with 3 dot's on the rest? Or something else?
In [103]: a = np.random.randn(30).reshape(3, 5, 2)
...: b = np.random.randn(30).reshape(3, 2, 5)
In [104]: (a#b).shape
Out[104]: (3, 5, 5)
In [105]: np.einsum('ijk,ikl->ijl',a,b).shape
Out[105]: (3, 5, 5)
#Ivan's answer is different:
In [106]: np.einsum('ijk,ilm->ijm', a, b).shape
Out[106]: (3, 5, 5)
In [107]: np.allclose(np.einsum('ijk,ilm->ijm', a, b), a#b)
Out[107]: False
In [108]: np.allclose(np.einsum('ijk,ikl->ijl', a, b), a#b)
Out[108]: True
Ivan's sums the k dimension of one, and l of the other, and then does a broadcasted elementwise. That is not matrix multiplication:
In [109]: (a.sum(axis=-1,keepdims=True)* b.sum(axis=1,keepdims=True)).shape
Out[109]: (3, 5, 5)
In [110]: np.allclose((a.sum(axis=-1,keepdims=True)* b.sum(axis=1,keepdims=True)),np.einsum('ijk,ilm->ijm', a,
...: b))
Out[110]: True
Another test of the batch processing:
In [112]: res=np.zeros((3,5,5))
...: for i in range(3):
...: res[i] = a[i]#b[i]
...: np.allclose(res, a#b)
Out[112]: True
Can someone explain in steps how numpy broadcasting works in this case?
a = np.ones((2,3))
b = np.ones((2,1,3))
c = a-b
a.shape
(2, 3)
b.shape
(2, 1, 3)
c.shape
(2, 2, 3)
Referring to this page, it says that numpy prepends the tensor with lower rank with 1s, so in this case we have
a.shape = [1,2,3]
Tile a along axis 1 to get a.shape=[2,2,3]
tile b along axis 2 to get b.shape=[2,2,3]
When the dimensions are same, subtract
Prepend 1 to a.shape, so a.shape -> (1,2,3)
Stretch a along dim 1 to match b. so a.shape -> (2,2,3)
Stretch b along dim 2 to match a. so b.shape -> (2,2,3)
Subtract
Is that what you're looking for?
I have an array A (shape = (a, 1)) and matrix B (shape = (b1, b2)). Want to multiply the latter by each element of the former to generate a tridimensional array (shape = (a, b1, b2)).
Is there a vectorized way to do this?
import numpy as np
A = np.random.rand(3, 1)
B = np.random.rand(5, 4)
C = np.array([ a * B for a in A ])
There are several ways you can achieve this.
One is using np.dot, note that it will be necessary to introduce a second axis in B so both ndarrays can be multiplied:
C = np.dot(A,B[:,None])
print(C.shape)
# (3, 5, 4)
Using np.multiply.outer, as #divakar suggests:
C = np.multiply.outer(A,B)
print(C.shape)
# (3, 5, 4)
Or you could also use np.einsum:
C = np.einsum('ij,kl->ikl', A, B)
print(C.shape)
# (3, 5, 4)