I'm using scikit.morphology to do an erosion on a two-dimensional array. I need to also ascertain the distance of each cell to the minimum value identified in the erosion.
Example:
np.reshape(np.arange(1,126,step=5),[5,5])
array([[ 1, 6, 11, 16, 21],
[ 26, 31, 36, 41, 46],
[ 51, 56, 61, 66, 71],
[ 76, 81, 86, 91, 96],
[101, 106, 111, 116, 121]])
erosion(np.reshape(np.arange(1,126,step=5),[5,5]),selem=disk(3))
array([[ 1, 1, 1, 1, 6],
[ 1, 1, 1, 6, 11],
[ 1, 1, 1, 6, 11],
[ 1, 6, 11, 16, 21],
[26, 31, 36, 41, 46]])
Now what I want to do is also return an array that gives me the distance to the minimum like this:
array([[ 0, 1, 2, 3, 3],
[ 1, 1, 2, 3, 3],
[ 2, 2, 3, 3, 3],
[ 3, 3, 3, 3, 3],
[ 3, 3, 3, 3, 3]])
Is there a scikit tool that can do this? If not, any tips on how to efficiently achieve this result?
You can find the distances from the centre of your footprint using scipy.ndimage.distance_transform_cdt, then use SciPy's ndimage.generic_filter to return those values:
import numpy as np
from skimage.morphology import erosion, disk
from scipy import ndimage as ndi
input_arr = np.reshape(np.arange(1,126,step=5),[5,5])
footprint = disk(3)
def distance_from_min(values, distance_values):
d = np.inf
min_val = np.inf
for i in range(len(values)):
if values[i] <= min_val:
min_val = values[i]
d = distance_values[i]
return d
full_footprint = np.ones_like(footprint, dtype=float)
full_footprint[tuple(i//2 for i in footprint.shape)] = 0
# use `ndi.distance_transform_edt` instead for the euclidean distance
distance_footprint = ndi.distance_transform_cdt(
full_footprint, metric='taxicab'
)
# set values outside footprint to 0 for pretty-printing
distance_footprint[~footprint.astype(bool)] = 0
# then, extract it into values matching the values in generic_filter
distance_values = distance_footprint[footprint.astype(bool)]
output = ndi.generic_filter(
input_arr.astype(float),
distance_from_min,
footprint=footprint,
mode='constant',
cval=np.inf,
extra_arguments=(distance_values,),
)
print('input:\n', input_arr)
print('footprint:\n', footprint)
print('distance_footprint:\n', distance_footprint)
print('output:\n', output)
Which gives:
input:
[[ 1 6 11 16 21]
[ 26 31 36 41 46]
[ 51 56 61 66 71]
[ 76 81 86 91 96]
[101 106 111 116 121]]
footprint:
[[0 0 0 1 0 0 0]
[0 1 1 1 1 1 0]
[0 1 1 1 1 1 0]
[1 1 1 1 1 1 1]
[0 1 1 1 1 1 0]
[0 1 1 1 1 1 0]
[0 0 0 1 0 0 0]]
distance_footprint:
[[0 0 0 3 0 0 0]
[0 4 3 2 3 4 0]
[0 3 2 1 2 3 0]
[3 2 1 0 1 2 3]
[0 3 2 1 2 3 0]
[0 4 3 2 3 4 0]
[0 0 0 3 0 0 0]]
output:
[[0. 1. 2. 3. 3.]
[1. 2. 3. 3. 3.]
[2. 3. 4. 4. 4.]
[3. 3. 3. 3. 3.]
[3. 3. 3. 3. 3.]]
This function will be very slow, however. If you want to make it faster, you will need (a) a solution like Numba or Cython for the filter function, in conjunction with SciPy LowLevelCallables and (b) to hardcode the distance array into the distance function, because for LowLevelCallables it is more difficult to pass in extra arguments. Here is a full example with llc-tools, which you can install with pip install numba llc-tools.
import numpy as np
from scipy import ndimage as ndi
from skimage.morphology import erosion, disk
import llc
def filter_func_from_footprint(footprint):
# first, create a footprint where the values are the distance from the
# center
full_footprint = np.ones_like(footprint, dtype=float)
full_footprint[tuple(i//2 for i in footprint.shape)] = 0
# use `ndi.distance_transform_edt` instead for the euclidean distance
distance_footprint = ndi.distance_transform_cdt(
full_footprint, metric='taxicab'
)
# then, extract it into values matching the values in generic_filter
distance_footprint[~footprint.astype(bool)] = 0
distance_values = distance_footprint[footprint.astype(bool)]
# finally, create a filter function with the values hardcoded
#llc.jit_filter_function
def distance_from_min(values):
d = np.inf
min_val = np.inf
for i in range(len(values)):
if values[i] <= min_val:
min_val = values[i]
d = distance_values[i]
return d
return distance_from_min
if __name__ == '__main__':
input_arr = np.reshape(np.arange(1,126,step=5),[5,5])
footprint = disk(3)
eroded = erosion(input_arr, selem=footprint)
filter_func = filter_func_from_footprint(footprint)
result = ndi.generic_filter(
# use input_arr.astype(float) when using euclidean dist
input_arr,
filter_func,
footprint=disk(3),
mode='constant',
cval=np.inf,
)
print('input:\n', input_arr)
print('output:\n', result)
Which gives:
input:
[[ 1 6 11 16 21]
[ 26 31 36 41 46]
[ 51 56 61 66 71]
[ 76 81 86 91 96]
[101 106 111 116 121]]
output:
[[0 1 2 3 3]
[1 2 3 3 3]
[2 3 4 4 4]
[3 3 3 3 3]
[3 3 3 3 3]]
For more reading on low-level callables and llc-tools, in addition to the LowLevelCallable documentation on the SciPy site (linked above, plus links therein), you can read these two blog posts I wrote a few years ago:
SciPy's new LowLevelCallable is a game-changer
Prettier LowLevelCallables with Numba JIT and decorators
Is there a way to slice a 2d array in numpy into smaller 2d arrays?
Example
[[1,2,3,4], -> [[1,2] [3,4]
[5,6,7,8]] [5,6] [7,8]]
So I basically want to cut down a 2x4 array into 2 2x2 arrays. Looking for a generic solution to be used on images.
There was another question a couple of months ago which clued me in to the idea of using reshape and swapaxes. The h//nrows makes sense since this keeps the first block's rows together. It also makes sense that you'll need nrows and ncols to be part of the shape. -1 tells reshape to fill in whatever number is necessary to make the reshape valid. Armed with the form of the solution, I just tried things until I found the formula that works.
You should be able to break your array into "blocks" using some combination of reshape and swapaxes:
def blockshaped(arr, nrows, ncols):
"""
Return an array of shape (n, nrows, ncols) where
n * nrows * ncols = arr.size
If arr is a 2D array, the returned array should look like n subblocks with
each subblock preserving the "physical" layout of arr.
"""
h, w = arr.shape
assert h % nrows == 0, f"{h} rows is not evenly divisible by {nrows}"
assert w % ncols == 0, f"{w} cols is not evenly divisible by {ncols}"
return (arr.reshape(h//nrows, nrows, -1, ncols)
.swapaxes(1,2)
.reshape(-1, nrows, ncols))
turns c
np.random.seed(365)
c = np.arange(24).reshape((4, 6))
print(c)
[out]:
[[ 0 1 2 3 4 5]
[ 6 7 8 9 10 11]
[12 13 14 15 16 17]
[18 19 20 21 22 23]]
into
print(blockshaped(c, 2, 3))
[out]:
[[[ 0 1 2]
[ 6 7 8]]
[[ 3 4 5]
[ 9 10 11]]
[[12 13 14]
[18 19 20]]
[[15 16 17]
[21 22 23]]]
I've posted an inverse function, unblockshaped, here, and an N-dimensional generalization here. The generalization gives a little more insight into the reasoning behind this algorithm.
Note that there is also superbatfish's
blockwise_view. It arranges the
blocks in a different format (using more axes) but it has the advantage of (1)
always returning a view and (2) being capable of handling arrays of any
dimension.
It seems to me that this is a task for numpy.split or some variant.
e.g.
a = np.arange(30).reshape([5,6]) #a.shape = (5,6)
a1 = np.split(a,3,axis=1)
#'a1' is a list of 3 arrays of shape (5,2)
a2 = np.split(a, [2,4])
#'a2' is a list of three arrays of shape (2,5), (2,5), (1,5)
If you have a NxN image you can create, e.g., a list of 2 NxN/2 subimages, and then divide them along the other axis.
numpy.hsplit and numpy.vsplit are also available.
There are some other answers that seem well-suited for your specific case already, but your question piqued my interest in the possibility of a memory-efficient solution usable up to the maximum number of dimensions that numpy supports, and I ended up spending most of the afternoon coming up with possible method. (The method itself is relatively simple, it's just that I still haven't used most of the really fancy features that numpy supports so most of the time was spent researching to see what numpy had available and how much it could do so that I didn't have to do it.)
def blockgen(array, bpa):
"""Creates a generator that yields multidimensional blocks from the given
array(_like); bpa is an array_like consisting of the number of blocks per axis
(minimum of 1, must be a divisor of the corresponding axis size of array). As
the blocks are selected using normal numpy slicing, they will be views rather
than copies; this is good for very large multidimensional arrays that are being
blocked, and for very large blocks, but it also means that the result must be
copied if it is to be modified (unless modifying the original data as well is
intended)."""
bpa = np.asarray(bpa) # in case bpa wasn't already an ndarray
# parameter checking
if array.ndim != bpa.size: # bpa doesn't match array dimensionality
raise ValueError("Size of bpa must be equal to the array dimensionality.")
if (bpa.dtype != np.int # bpa must be all integers
or (bpa < 1).any() # all values in bpa must be >= 1
or (array.shape % bpa).any()): # % != 0 means not evenly divisible
raise ValueError("bpa ({0}) must consist of nonzero positive integers "
"that evenly divide the corresponding array axis "
"size".format(bpa))
# generate block edge indices
rgen = (np.r_[:array.shape[i]+1:array.shape[i]//blk_n]
for i, blk_n in enumerate(bpa))
# build slice sequences for each axis (unfortunately broadcasting
# can't be used to make the items easy to operate over
c = [[np.s_[i:j] for i, j in zip(r[:-1], r[1:])] for r in rgen]
# Now to get the blocks; this is slightly less efficient than it could be
# because numpy doesn't like jagged arrays and I didn't feel like writing
# a ufunc for it.
for idxs in np.ndindex(*bpa):
blockbounds = tuple(c[j][idxs[j]] for j in range(bpa.size))
yield array[blockbounds]
You question practically the same as this one. You can use the one-liner with np.ndindex() and reshape():
def cutter(a, r, c):
lenr = a.shape[0]/r
lenc = a.shape[1]/c
np.array([a[i*r:(i+1)*r,j*c:(j+1)*c] for (i,j) in np.ndindex(lenr,lenc)]).reshape(lenr,lenc,r,c)
To create the result you want:
a = np.arange(1,9).reshape(2,1)
#array([[1, 2, 3, 4],
# [5, 6, 7, 8]])
cutter( a, 1, 2 )
#array([[[[1, 2]],
# [[3, 4]]],
# [[[5, 6]],
# [[7, 8]]]])
Some minor enhancement to TheMeaningfulEngineer's answer that handles the case when the big 2d array cannot be perfectly sliced into equally sized subarrays
def blockfy(a, p, q):
'''
Divides array a into subarrays of size p-by-q
p: block row size
q: block column size
'''
m = a.shape[0] #image row size
n = a.shape[1] #image column size
# pad array with NaNs so it can be divided by p row-wise and by q column-wise
bpr = ((m-1)//p + 1) #blocks per row
bpc = ((n-1)//q + 1) #blocks per column
M = p * bpr
N = q * bpc
A = np.nan* np.ones([M,N])
A[:a.shape[0],:a.shape[1]] = a
block_list = []
previous_row = 0
for row_block in range(bpc):
previous_row = row_block * p
previous_column = 0
for column_block in range(bpr):
previous_column = column_block * q
block = A[previous_row:previous_row+p, previous_column:previous_column+q]
# remove nan columns and nan rows
nan_cols = np.all(np.isnan(block), axis=0)
block = block[:, ~nan_cols]
nan_rows = np.all(np.isnan(block), axis=1)
block = block[~nan_rows, :]
## append
if block.size:
block_list.append(block)
return block_list
Examples:
a = np.arange(25)
a = a.reshape((5,5))
out = blockfy(a, 2, 3)
a->
array([[ 0, 1, 2, 3, 4],
[ 5, 6, 7, 8, 9],
[10, 11, 12, 13, 14],
[15, 16, 17, 18, 19],
[20, 21, 22, 23, 24]])
out[0] ->
array([[0., 1., 2.],
[5., 6., 7.]])
out[1]->
array([[3., 4.],
[8., 9.]])
out[-1]->
array([[23., 24.]])
For now it just works when the big 2d array can be perfectly sliced into equally sized subarrays.
The code bellow slices
a ->array([[ 0, 1, 2, 3, 4, 5],
[ 6, 7, 8, 9, 10, 11],
[12, 13, 14, 15, 16, 17],
[18, 19, 20, 21, 22, 23]])
into this
block_array->
array([[[ 0, 1, 2],
[ 6, 7, 8]],
[[ 3, 4, 5],
[ 9, 10, 11]],
[[12, 13, 14],
[18, 19, 20]],
[[15, 16, 17],
[21, 22, 23]]])
p ang q determine the block size
Code
a = arange(24)
a = a.reshape((4,6))
m = a.shape[0] #image row size
n = a.shape[1] #image column size
p = 2 #block row size
q = 3 #block column size
block_array = []
previous_row = 0
for row_block in range(blocks_per_row):
previous_row = row_block * p
previous_column = 0
for column_block in range(blocks_per_column):
previous_column = column_block * q
block = a[previous_row:previous_row+p,previous_column:previous_column+q]
block_array.append(block)
block_array = array(block_array)
If you want a solution that also handles the cases when the matrix is
not equally divided, you can use this:
from operator import add
half_split = np.array_split(input, 2)
res = map(lambda x: np.array_split(x, 2, axis=1), half_split)
res = reduce(add, res)
Here is a solution based on unutbu's answer that handle case where matrix cannot be equally divided. In this case, it will resize the matrix before using some interpolation. You need OpenCV for this. Note that I had to swap ncols and nrows to make it works, didn't figured why.
import numpy as np
import cv2
import math
def blockshaped(arr, r_nbrs, c_nbrs, interp=cv2.INTER_LINEAR):
"""
arr a 2D array, typically an image
r_nbrs numbers of rows
r_cols numbers of cols
"""
arr_h, arr_w = arr.shape
size_w = int( math.floor(arr_w // c_nbrs) * c_nbrs )
size_h = int( math.floor(arr_h // r_nbrs) * r_nbrs )
if size_w != arr_w or size_h != arr_h:
arr = cv2.resize(arr, (size_w, size_h), interpolation=interp)
nrows = int(size_w // r_nbrs)
ncols = int(size_h // c_nbrs)
return (arr.reshape(r_nbrs, ncols, -1, nrows)
.swapaxes(1,2)
.reshape(-1, ncols, nrows))
a = np.random.randint(1, 9, size=(9,9))
out = [np.hsplit(x, 3) for x in np.vsplit(a,3)]
print(a)
print(out)
yields
[[7 6 2 4 4 2 5 2 3]
[2 3 7 6 8 8 2 6 2]
[4 1 3 1 3 8 1 3 7]
[6 1 1 5 7 2 1 5 8]
[8 8 7 6 6 1 8 8 4]
[6 1 8 2 1 4 5 1 8]
[7 3 4 2 5 6 1 2 7]
[4 6 7 5 8 2 8 2 8]
[6 6 5 5 6 1 2 6 4]]
[[array([[7, 6, 2],
[2, 3, 7],
[4, 1, 3]]), array([[4, 4, 2],
[6, 8, 8],
[1, 3, 8]]), array([[5, 2, 3],
[2, 6, 2],
[1, 3, 7]])], [array([[6, 1, 1],
[8, 8, 7],
[6, 1, 8]]), array([[5, 7, 2],
[6, 6, 1],
[2, 1, 4]]), array([[1, 5, 8],
[8, 8, 4],
[5, 1, 8]])], [array([[7, 3, 4],
[4, 6, 7],
[6, 6, 5]]), array([[2, 5, 6],
[5, 8, 2],
[5, 6, 1]]), array([[1, 2, 7],
[8, 2, 8],
[2, 6, 4]])]]
I publish my solution. Notice that this code doesn't' actually create copies of original array, so it works well with big data. Moreover, it doesn't crash if array cannot be divided evenly (but you can easly add condition for that by deleting ceil and checking if v_slices and h_slices are divided without rest).
import numpy as np
from math import ceil
a = np.arange(9).reshape(3, 3)
p, q = 2, 2
width, height = a.shape
v_slices = ceil(width / p)
h_slices = ceil(height / q)
for h in range(h_slices):
for v in range(v_slices):
block = a[h * p : h * p + p, v * q : v * q + q]
# do something with a block
This code changes (or, more precisely, gives you direct access to part of an array) this:
[[0 1 2]
[3 4 5]
[6 7 8]]
Into this:
[[0 1]
[3 4]]
[[2]
[5]]
[[6 7]]
[[8]]
If you need actual copies, Aenaon code is what you are looking for.
If you are sure that big array can be divided evenly, you can use numpy splitting tools.
to add to #Aenaon answer and his blockfy function, if you are working with COLOR IMAGES/ 3D ARRAY here is my pipeline to create crops of 224 x 224 for 3 channel input
def blockfy(a, p, q):
'''
Divides array a into subarrays of size p-by-q
p: block row size
q: block column size
'''
m = a.shape[0] #image row size
n = a.shape[1] #image column size
# pad array with NaNs so it can be divided by p row-wise and by q column-wise
bpr = ((m-1)//p + 1) #blocks per row
bpc = ((n-1)//q + 1) #blocks per column
M = p * bpr
N = q * bpc
A = np.nan* np.ones([M,N])
A[:a.shape[0],:a.shape[1]] = a
block_list = []
previous_row = 0
for row_block in range(bpc):
previous_row = row_block * p
previous_column = 0
for column_block in range(bpr):
previous_column = column_block * q
block = A[previous_row:previous_row+p, previous_column:previous_column+q]
# remove nan columns and nan rows
nan_cols = np.all(np.isnan(block), axis=0)
block = block[:, ~nan_cols]
nan_rows = np.all(np.isnan(block), axis=1)
block = block[~nan_rows, :]
## append
if block.size:
block_list.append(block)
return block_list
then extended above to
for file in os.listdir(path_to_crop): ### list files in your folder
img = io.imread(path_to_crop + file, as_gray=False) ### open image
r = blockfy(img[:,:,0],224,224) ### crop blocks of 224 x 224 for red channel
g = blockfy(img[:,:,1],224,224) ### crop blocks of 224 x 224 for green channel
b = blockfy(img[:,:,2],224,224) ### crop blocks of 224 x 224 for blue channel
for x in range(0,len(r)):
img = np.array((r[x],g[x],b[x])) ### combine each channel into one patch by patch
img = img.astype(np.uint8) ### cast back to proper integers
img_swap = img.swapaxes(0, 2) ### need to swap axes due to the way things were proceesed
img_swap_2 = img_swap.swapaxes(0, 1) ### do it again
Image.fromarray(img_swap_2).save(path_save_crop+str(x)+"bounding" + file,
format = 'jpeg',
subsampling=0,
quality=100) ### save patch with new name etc
I want to do some forces calculations between vertices and because the forces are symmetrical I have a list of vertice-pairs that need those forces added. I am sure it's possible with fancy indexing, but I really just can get it to work with a slow python for-loop. for symmetric reasons, the right-hand side of the index array needs a negative sign when adding the forces.
consider you have the vertice index array:
>>> I = np.array([[0,1],[1,2],[2,0]])
I = [[0 1]
[1 2]
[2 0]]
and the x,y forces array for each pair:
>>> F = np.array([[3,6],[4,7],[5,8]])
F = [[3 6]
[4 7]
[5 8]]
the wanted operation could be described as:
"vertice #0 sums the force vectors (3,6) and (-5,-8),
vertice #1 sums the force vectors (-3,-6) and (4,7),
vertice #2 sums the force vectors (-4,-7) and (5,8)"
Desired results:
[ 3 6 ] [ 0 0 ] [-5 -8 ] [-2 -2 ] //resulting force Vertice #0
A = [-3 -6 ] + [ 4 7 ] + [ 0 0 ] = [ 1 1 ] //resulting force Vertice #1
[ 0 0 ] [-4 -7 ] [ 5 8 ] [ 1 1 ] //resulting force Vertice #2
edit:
my ugly for-loop solution:
import numpy as np
I = np.array([[0,1],[1,2],[2,0]])
F = np.array([[3,6],[4,7],[5,8]])
A = np.zeros((3,2))
A_x = np.zeros((3,2))
A_y = np.zeros((3,2))
for row in range(0,len(F)):
A_x[I[row][0],0]= F[row][0]
A_x[I[row][1],1]= -F[row][0]
A_y[I[row][0],0]= F[row][1]
A_y[I[row][1],1]= -F[row][1]
A = np.hstack((np.sum(A_x,axis=1).reshape((3,1)),np.sum(A_y,axis=1).reshape((3,1))))
print(A)
A= [[-2. -2.]
[ 1. 1.]
[ 1. 1.]]
Your current "push-style" interpretation of I is
For row-index k in I, take the forces from F[k] and add/subtract them to out[I[k], :]
I = np.array([[0,1],[1,2],[2,0]])
out = numpy.zeros_like(F)
for k, d in enumerate(I):
out[d[0], :] += F[k]
out[d[1], :] -= F[k]
out
# array([[-2, -2],
# [ 1, 1],
# [ 1, 1]])
However you can also change the meaning of I on its head and make it "pull-style", so it says
For row-index k in I, set vertex out[k] to be the difference of F[I[k]]
I = np.array([[0,2],[1,0],[2,1]])
out = numpy.zeros_like(F)
for k, d in enumerate(I):
out[k, :] = F[d[0], :] - F[d[1], :]
out
# array([[-2, -2],
# [ 1, 1],
# [ 1, 1]])
In which case the operation simplifies quite easily to mere fancy indexing:
out = F[I[:, 0], :] - F[I[:, 1], :]
# array([[-2, -2],
# [ 1, 1],
# [ 1, 1]])
You can preallocate an array to hold the shuffled forces and then use the index like so:
>>> N = I.max() + 1
>>> out = np.zeros((N, 2, 2), F.dtype)
>>> out[I, [1, 0]] = F[:, None, :]
>>> np.diff(out, axis=1).squeeze()
array([[-2, -2],
[ 1, 1],
[ 1, 1]])
or, equivalently,
>>> out = np.zeros((2, N, 2), F.dtype)
>>> out[[[1], [0]], I.T] = F
>>> np.diff(out, axis=0).squeeze()
array([[-2, -2],
[ 1, 1],
[ 1, 1]])
The way I understand the question, the values in the I array represent the vortex number, or the name of the vortex. They are not an actual positional index. Based on this thought, I have a different solution that uses the original I array. It does not quite come without loops, but should be OK for a reasonable number of vertices:
I = np.array([[0,1],[1,2],[2,0]])
F = np.array([[3,6],[4,7],[5,8]])
pos = I[:, 0]
neg = I[:, 1]
A = np.zeros_like(F)
unique = np.unique(I)
for i, vortex_number in enumerate(unique):
A[i] = F[np.where(pos==vortex_number)] - F[np.where(neg==vortex_number)]
# produces the expected result
# [[-2 -2]
# [ 1 1]
# [ 1 1]]
Maybe this loop can also be replaced by some numpy magic.