问题
One parameter in my model controls the dimension of arrays; it can vary from 1 to any positive integer: for my purposes this can be up to 20.
The flow of the program goes through a number of loops that depend on this dimension.
For example, if the value of the parameter is one I would have:
for i1 in range(0,100,1):
do stuff
If the value of the parameter is two, then I would have something like:
for i1 in range (0,100,1):
for i2 in range (0,100,1):
do stuff
Or, if the value of the parameter is three I would have:
for i1 in range (0,100,1):
for i2 in range (0,100,1):
for i3 in range (0,100,1):
do stuff
The dimension can change, so it's not possible to specify in advance how many nested loops will be needed; this has to written in some parametric way.
回答1:
Since you've listed no processing in the intermediate loops -- only in the innermost loop, I feel that what you really need is an iterator for your sequence of indices:
Let max_dim
be the dimensionality of your space, the quantity of dimensions.
max_val = 100
one_dim = list(range(max_val))
all_dim = [one_dim] * max_val
all_dim
is now a list of lists, one for each dimension. Each list contains the values 0-99, the very values your nested loops are using. Now for the magic steps from itertools
:
from itertools import product
for index_list in product(*all_dim):
# do your stuff; the index_list is [i1, i2, i3, ...]
This will iterate through your desired dimensions. For a small example, here's how the product
sequence looks with only two values and three dimensions:
>>> all_dim = [[0,1]] * 3
>>> all_dim
[[0, 1], [0, 1], [0, 1]]
>>> list(product(*all_dim))
[(0, 0, 0), (0, 0, 1), (0, 1, 0), (0, 1, 1), (1, 0, 0), (1, 0, 1), (1, 1, 0), (1, 1, 1)]
Does that handle your problem well enough?
回答2:
Use recursion for variable depth nested loops.
def nested(parameters, depth):
for x in range(0,100,1):
if depth==0:
do
else:
nested(parameters, depth-1)
来源:https://stackoverflow.com/questions/52839094/parametric-nested-loops-in-python