Is variable in Julia thread for loop thread local? - multithreading

Threads.#threads for i in [sth]
x = 123
...
end
Is x thread local. I can not find any document about it.
If not, how can I get a thread local one.

Assuming that x is local inside the loop then yes. Here is an example how you can check it:
# julia -t 2 --banner=no
julia> Threads.#threads for i in 1:8
x = i
#info Threads.threadid(), i, x
sleep(rand())
#info Threads.threadid(), i, x
end
[ Info: (1, 1, 1)
[ Info: (2, 5, 5)
[ Info: (1, 1, 1)
[ Info: (1, 2, 2)
[ Info: (2, 5, 5)
[ Info: (2, 6, 6)
[ Info: (1, 2, 2)
[ Info: (1, 3, 3)
[ Info: (2, 6, 6)
[ Info: (1, 3, 3)
[ Info: (1, 4, 4)
[ Info: (2, 7, 7)
[ Info: (1, 4, 4)
[ Info: (2, 7, 7)
[ Info: (2, 8, 8)
[ Info: (2, 8, 8)
For each thread whatever is passed inside the for loop body that is annotated with Threads.#threads is wrapped in an anonymous function. You can check it by using #macroexpand function.
Note that it is even more x: gets a new binding with each iteration of the loop (this is a standard behavior of Julia -- no matter if it is single or multi threaded code, see here).
If x is not a local variable inside the for loop this is not true any more as you can see here:
# julia -t 2 --banner=no
julia> function f()
x = 1
Threads.#threads for i in 1:8
x = i
#info Threads.threadid(), i, x
sleep(rand())
#info Threads.threadid(), i, x
end
end
f (generic function with 1 method)
julia> f()
[ Info: (1, 1, 5)
[ Info: (2, 5, 5)
[ Info: (1, 1, 5)
[ Info: (1, 2, 2)
[ Info: (2, 5, 2)
[ Info: (2, 6, 6)
[ Info: (2, 6, 6)
[ Info: (2, 7, 7)
[ Info: (1, 2, 7)
[ Info: (1, 3, 3)
[ Info: (1, 3, 3)
[ Info: (1, 4, 4)
[ Info: (2, 7, 4)
[ Info: (2, 8, 8)
[ Info: (2, 8, 8)
[ Info: (1, 4, 8)

Related

incorrect result showing in K nearest neighbour approach

I am reforming the 2D coordinate number in a aligned way which was not aligned (coordinate numbers were suffled) before.
I have below input coordinates,
X = [2, 2, 3, 4, 4, 4, 4, 5, 6, 6, 6, 6, 6, 5, 4, 3, 5, 5, 5]
Y = [2, 3, 3, 3, 4, 5, 6, 6, 6, 5, 4, 3, 2, 2, 2, 2, 3, 4, 5]
I have to make it aligned. Therefore, I first applied Sorted function on this coordinates. I got below output after it.
merged_list1 = sorted(zip(X, Y))
output
X1_coordinate_reformed = [2, 2, 3, 3, 4, 4, 4, 4, 4, 5, 5, 5, 5, 5, 6, 6, 6, 6, 6]
Y1_coordinate_reformed = [2, 3, 2, 3, 2, 3, 4, 5, 6, 2, 3, 4, 5, 6, 2, 3, 4, 5, 6]
Still it iot aligned properly. I want two consecutive nodes place next to each other. Therefore I am applying the approach to find the nearest coordinate from origin to find the very first node. Then from the first node, I found another nearest coordinate and so on...For that, I have applied below code,
First I wrote a function which calculates the distance and gives index of the nearest coordinate from the list.
def solve(pts, pt):
x, y = pt
idx = -1
smallest = float("inf")
for p in pts:
if p[0] == x or p[1] == y:
dist = abs(x - p[0]) + abs(y - p[1])
if dist < smallest:
idx = pts.index(p)
smallest = dist
elif dist == smallest:
if pts.index(p) < idx:
idx = pts.index(p)
smallest = dist
return idx
coor2 = list(zip(X1_coordinate_reformed, Y1_coordinate_reformed)) # make a list which contains tuples of X and Y coordinates
pts2 = coor2.copy()
origin1 = (0, 0)
new_coor1 = []
for i in range(len(pts2)):
pt = origin1
index_num1 = solve(pts2, pt)
print('index is', index_num1)
origin1 = pts2[index_num1]
new_coor1.append(pts2[index_num1])
del pts2[index_num1]
After running the code, I got below output,
[(6, 6), (5, 6), (4, 6), (4, 5), (4, 4), (4, 3), (3, 3), (2, 3), (2, 2), (3, 2), (4, 2), (5, 2), (5, 3), (5, 4), (5, 5), (6, 5), (6, 4), (6, 3), (6, 2)]
Which is not correct because it can be clearly understand that,
coor2 = [(2, 2), (2, 3), (3, 2), (3, 3), (4, 2), (4, 3), (4, 4), (4, 5), (4, 6), (5, 2), (5, 3), (5, 4), (5, 5), (5, 6), (6, 2), (6, 3), (6, 4), (6, 5), (6, 6)]
origin = (0, 0)
if we find the distance between Origin which was (0, 0) in very first and from every coordinate from above coor2 list, we will get (2,2) is nearest coordinate. Then How come my code gives (6,6) is the nearest coordinate??
The interesting thing is, if I apply the same procedure (sorting followed by finding nearest coordinate) on below coordinates,
X2_coordinate = [2, 4, 4, 2, 3, 2, 4, 3, 1, 3, 4, 3, 1, 2, 0, 3, 4, 2, 0]
Y2_coordinate = [3, 4, 2, 1, 3, 2, 1, 0, 0, 2, 3, 4, 1, 4, 0, 1, 0, 0, 1]
After applying sorted function
X2_coordinate_reformed = [0, 0, 1, 1, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 4, 4, 4, 4, 4]
Y2_coordinate_reformed = [0, 1, 0, 1, 0, 1, 2, 3, 4, 0, 1, 2, 3, 4, 0, 1, 2, 3, 4]
After applying method of searching nearest coordinates mentioned above, the result I got
[(0, 0), (0, 1), (1, 1), (1, 0), (2, 0), (2, 1), (2, 2), (2, 3), (2, 4), (3, 4), (3, 3), (3, 2), (3, 1), (3, 0), (4, 0), (4, 1), (4, 2), (4, 3), (4, 4)]
Kindly suggest me where I am doing wrong and what should I change??
It is better to use scipy for finding closest coordinate.
The code given below works.
from scipy import spatial
pts = merged_list1.copy()
origin = (0, 0)
origin = np.array(origin)
new_coordi = []
for i in range(len(pts)):
x = origin
distance,index = spatial.KDTree(pts).query(x)
new_coordi.append(pts[index])
origin = np.array(pts[index])
del pts[index]

multiprocessing gives wrong number of outputs, repeats outside scope

When I use multiprocessing.pool.starmap it repeats the process outside the part I define as the process.
I am confused as to why.
MRA:
from multiprocessing import Pool
def countAdd( first, second, third ):
added = str( first + second + third )
with open( "test.txt", 'a' ) as f:
f.write( "this is string: " + added + '\n')
testlist = [ 1, 2, 3, 4, 5, 6, ]
testlist2 = [ 7, 8, 9, 10, 11, 12 ]
testlist3 = [ 13, 14, 15, 16, 17, 18 ]
arglist = [ (testlist[ test ], testlist2[ test ], testlist3[ test ]) for test in range( 0, len( testlist ) ) ]
print( arglist )
print( len( testlist ) )
if __name__ == '__main__':
with Pool( 3 ) as pool:
pool.starmap( countAdd, arglist )
Output:
[(1, 7, 13), (2, 8, 14), (3, 9, 15), (4, 10, 16), (5, 11, 17), (6, 12, 18)]
6
[(1, 7, 13), (2, 8, 14), (3, 9, 15), (4, 10, 16), (5, 11, 17), (6, 12, 18)]
6
[(1, 7, 13), (2, 8, 14), (3, 9, 15), (4, 10, 16), (5, 11, 17), (6, 12, 18)]
6
[(1, 7, 13), (2, 8, 14), (3, 9, 15), (4, 10, 16), (5, 11, 17), (6, 12, 18)]
6
file:
this is string: 21
this is string: 24
this is string: 27
this is string: 33
this is string: 36
So I'm really confused. Why are the print statements also printed repeatedly? But only 4 times?!
Why does the file only contain 5 times the output of the function and not 6 times, eventhough the arglist contains 6 tuples?
I searched around, found a similar probleme but isn't it.
Python Multiprocessing data output wrong
Thanks for taking the time to respond.
EDIT:
Based on the comments by MisterMiyagi I changed my code to the following, but there are still problems:
from multiprocessing import Pool
def countAdd( first, second, third ):
added = str( first + second + third )
with open( "test.txt", 'a' ) as f:
f.write( "this is string: " + added + '\n')
if __name__ == '__main__':
testlist = [ 1, 2, 3, 4, 5, 6, ]
testlist2 = [ 7, 8, 9, 10, 11, 12 ]
testlist3 = [ 13, 14, 15, 16, 17, 18 ]
arglist = [ (testlist[ test ], testlist2[ test ], testlist3[ test ]) for test in range( 0, len( testlist ) ) ]
print( arglist )
print( len( arglist ) )
with Pool( 3 ) as pool:
pool.starmap( countAdd, arglist )
print( "done")
I repeatedly ran the code. Sometimes I get the six lines of output but othertimes I don't. I ran the code 10 times and I got 54 lines of text in the output file. What is going on here?

How to reshape array of shape (2, *(x, y)) to (1, *(x,y), 2) while preserving the values of (x, y)?

I want to reshape array of shape (2, *(x, y)) to (1, *(x,y), 2) while preserving the values of (x, y)?
(2, *(x,y)) where 2 represents the frames of game screen with (x, y) being an array with pixel values. I wish to convert it into an array of shape of (1, *(x, y), 2), such that the number 2 still represents the frame index, while (x,y) array value is preserved. 1 will be used to index the batch for training the neural network.
numpy.reshape(1, *(x,y), 2) doesn't preserve the (x,y) array.
Use numpy.transpose(), e.g.:
import numpy as np
arr = np.arange(2 * 3 * 4).reshape((2, 3, 4))
arr.shape
# (2, 3, 4)
arr.transpose(1, 2, 0).shape
# (3, 4, 2)
new_arr = arr.transpose(1, 2, 0)[None, ...]
new_arr.shape
# (1, 3, 4, 2)
# the `(3, 4)` array is preserved:
arr.transpose(1, 2, 0)[:, :, 0]
# array([[ 0, 1, 2, 3],
# [ 4, 5, 6, 7],
# [ 8, 9, 10, 11]])
arr[0, :, :]
# array([[ 0, 1, 2, 3],
# [ 4, 5, 6, 7],
# [ 8, 9, 10, 11]])

How to assign a random weight to edges in networkx, like weight of edge (a, a) = 0 and weight of edge (a, b) = K, where K is some random number

I am working on weighted graphs and I would like to assign a random weight for the edges of the graph, such that,
weight of edge(a, a) = 0
weight of (a, b) = weight of edge(b, a) = K
where K is some random number. This goes on for all the edges of the graphs.
For that, I am using random.randint() method. I am actually using the logic of sum. If sum of both the edges is same, then assign some random integer.
Here is my code,
nodelist = list(range(1, num_nodes + 1))
edgelist = []
for i in nodelist:
for j in nodelist:
if i == j:
edgelist.append((i, j, 0))
if (i != j and sum((i, j)) == sum((j, i))):
rand = random.randint(5, 25)
edgelist.append((i, j, rand))
print(edgelist)
Actual result,
[(1, 1, 0), (1, 2, 18), (1, 3, 6), (2, 1, 13), (2, 2, 0), (2, 3, 21), (3, 1, 20), (3, 2, 17), (3, 3, 0)]
Expected result,
[(1, 1, 0), (1, 2, K), (1, 3, H), (2, 1, K), (2, 2, 0), (2, 3, P), (3, 1, H), (3, 2, P), (3, 3, 0)]
where, K, H, P are some random integers.
If the ordering of the result is not important following code gives the desired output:
import random
num_nodes = 3
nodelist = list(range(1, num_nodes + 1))
edgelist = []
for i in nodelist:
for j in nodelist:
if j > i:
break
if i == j:
edgelist.append((i, j, 0))
else:
rand = random.randint(5, 25)
edgelist.append((i, j, rand))
edgelist.append((j, i, rand))
print(edgelist)
# [(1, 1, 0), (2, 1, 7), (1, 2, 7), (2, 2, 0), (3, 1, 18), (1, 3, 18), (3, 2, 13), (2, 3, 13), (3, 3, 0)]
In case you need the edges sorted, simply use:
print(sorted(edgelist))
# [(1, 1, 0), (1, 2, 20), (1, 3, 16), (2, 1, 20), (2, 2, 0), (2, 3, 23), (3, 1, 16), (3, 2, 23), (3, 3, 0)]
Just a little change in your code will do the trick.
Here is the solution I found to obtain your expected output
num_nodes = 3
nodelist = list(range(1, num_nodes + 1))
edgelist = []
for i in nodelist:
for j in nodelist:
if i == j:
edgelist.append((i, j, 0))
elif i < j:
rand = random.randint(5, 25)
edgelist.append((i, j, rand))
edgelist.append((j, i, rand))
print(sorted(edgelist))
This code outputs :
[(1, 1, 0), (1, 2, 15), (1, 3, 15), (2, 1, 15), (2, 2, 0), (2, 3, 21), (3, 1, 15), (3, 2, 21), (3, 3, 0)]
So I figured out something interesting. Say below matrix shows edges in a complete graph of 5 nodes,
[1, 1] [1, 2] [1, 3] [1, 4] [1, 5]
[2, 1] [2, 2] [2, 3] [2, 4] [2, 5]
[3, 1] [3, 2] [3, 3] [3, 4] [3, 5]
[4, 1] [4, 2] [4, 3] [4, 4] [4, 5]
[5, 1] [5, 2] [5, 3] [5, 4] [5, 5]
now, moving right side from principal diagonal, we have lists whose first element is less than second element. We just got to target them and append new random weight to it.
Here is my code,
nodelist = list(range(1, num_nodes + 1))
edgelist = []
for i in nodelist:
for j in nodelist:
edgelist.append([i, j])
p = 0
eff_edgelist = []
while p < len(edgelist):
if edgelist[p][0] <= edgelist[p][1]:
eff_edgelist.append(edgelist[p])
p += 1
for i in eff_edgelist:
if i[0] == i[1]:
i.append(0)
else:
i.append(random.randint(5, 50))
eff_edgelist = [tuple(i) for i in eff_edgelist]
for i in list(G.edges(data=True)):
print([i])
and the result,
[(1, 1, {'weight': 0})]
[(1, 2, {'weight': 12})]
[(1, 3, {'weight': 37})]
[(1, 4, {'weight': 38})]
[(1, 5, {'weight': 6})]
[(2, 2, {'weight': 0})]
[(2, 3, {'weight': 12})]
[(2, 4, {'weight': 40})]
[(2, 5, {'weight': 8})]
[(3, 3, {'weight': 0})]
[(3, 4, {'weight': 15})]
[(3, 5, {'weight': 38})]
[(4, 4, {'weight': 0})]
[(4, 5, {'weight': 41})]
[(5, 5, {'weight': 0})]
and if you check, print(G[2][1]), the output will be {'weight': 12},
which means weight of edge(a, b) = weight of edge(b, a).

How to define functions from an index set to an indefinite set of domains?

I want to define sth like
list([[i0,i1,i2,i3, ..., ik]] for i0 in T[0] for i1 in T[1] for i2 in T[2] for i3 in T[3] for ...)
as k is indefinite, I cannot do this like
list([[i0,i1,i2,i3]] for i0 in T[0] for i1 in T[1] for i2 in T[2] for i3 in T[3]).
Is there a general solution?
Many thanks!
Your nested fors will make a Cartesian product of the sublists in T. Itertools has a product() function that will give you an iterator of these values, which you can use like:
from itertools import product
T = [[1, 2, 3], [4, 5, 6], [7, 8, 9], [10,11, 12]]
p = product(*T)
for i in p:
print(i)
(1, 4, 7, 10)
(1, 4, 7, 11)
(1, 4, 7, 12)
(1, 4, 8, 10)
(1, 4, 8, 11)
(1, 4, 8, 12)
(1, 4, 9, 10)
(1, 4, 9, 11)
...
(3, 6, 9, 10)
(3, 6, 9, 11)
(3, 6, 9, 12)
Of course you can also pass it to list() if want the values in a list.

Resources