Reputation: 122112
Other than the following brute force method, is there a simpler way to allocate value to a variable given a if condition?
Method 1:
a, b, c, d = 0.03,0.4,0.055,0.7
x = 0.2
if a < x:
a = x
if b < x:
b = x
if c < x:
c = x
if d < x:
d = x
Upvotes: 0
Views: 198
Reputation: 8400
Absolutely consider using numpy.where
which is the most efficient way to do what you want dealing with any size of array and dimension:
#your example:
a,b,c,d = 0.03,0.4,0.055,0.7
x = 0.2
#solution
values = numpy.asarray([a, b, c, d])
a,b,c,d = numpy.where(values<x, x, values)
#efficiency becomes clear when
values = numpy.random.rand(1000,100,10) #any size and number of dimensions
values = numpy.where(values<x, x, values) #just works fine and efficient
#further developments would be possible, e.g., multiple conditions
values = numpy.where((values>=0.3)&(values<0.7), 0.5, values)
Upvotes: 2
Reputation: 161
Perhaps, a bit more Haskell-like (zipWith)
from itertools import izip, starmap, repeat
a, b, c, d = starmap(max, izip(repeat(0.2), (0.03, 0.4, 0.055, 0.7)))
Some basic timeits (darwin 12.2.0, py 2.7.3):
In [0]: %timeit a,b,c,d = starmap(max, izip(repeat(0.2), (0.03, 0.4, 0.055, 0.7)))
1000000 loops, best of 3: 1.87 us per loop
In [1]: %timeit a,b,c,d = map(max, izip(repeat(0.2), (0.03, 0.4, 0.055, 0.7)))
100000 loops, best of 3: 3.99 us per loop
In [2]: %timeit a,b,c,d = [max(0.2, v) for v in [0.03,0.4,0.055,0.7]]
100000 loops, best of 3: 1.95 us per loop
In [3]: %timeit a,b,c,d = [max(0.2, v) for v in (0.03,0.4,0.055,0.7)]
1000000 loops, best of 3: 1.62 us per loop
Conclusions:
Tuples are faster to iterate over than lists ?!?
starmap beats map, even though max(values) is faster than max(*values) ?!?
Upvotes: 0
Reputation: 5736
try:
a = x if a < x else a
b = x if b < x else b
c = x if c < x else c
d = x if d < x else d
Upvotes: -3
Reputation: 95682
Perhaps:
a, b, c, d = max(a, x), max(b, x), max(c, x), max(d, x)
but if you have a lot of variables being handled in exactly the same way a list
might be better.
values = [0.03,0.4,0.055,0.7]
x = 0.2
values = [max(v, x) for v in values]
Upvotes: 20