How does reduce function work? - python

As far as I understand, the reduce function takes a list l and a function f. Then, it calls the function f on first two elements of the list and then repeatedly calls the function f with the next list element and the previous result.
So, I define the following functions:
The following function computes the factorial.
def fact(n):
if n == 0 or n == 1:
return 1
return fact(n-1) * n
def reduce_func(x,y):
return fact(x) * fact(y)
lst = [1, 3, 1]
print reduce(reduce_func, lst)
Now, shouldn't this give me ((1! * 3!) * 1!) = 6? But, instead it gives 720. Why 720? It seems to take the factorial of 6 too. But, I need to understand why.
Can someone explains why this happens and a work-around?
I basically want to compute the product of factorials of all the entries in the list.
The backup plan is to run a loop and compute it. But, I would prefer using reduce.

The other answers are great. I'll simply add an illustrated example that I find pretty good to understand reduce():
>>> reduce(lambda x,y: x+y, [47,11,42,13])
113
will be computed as follows:
(Source) (mirror)

The easiest way to understand reduce() is to look at its pure Python equivalent code:
def myreduce(func, iterable, start=None):
it = iter(iterable)
if start is None:
try:
start = next(it)
except StopIteration:
raise TypeError('reduce() of empty sequence with no initial value')
accum_value = start
for x in iterable:
accum_value = func(accum_value, x)
return accum_value
You can see that it only makes sense for your reduce_func() to apply the factorial to the rightmost argument:
def fact(n):
if n == 0 or n == 1:
return 1
return fact(n-1) * n
def reduce_func(x,y):
return x * fact(y)
lst = [1, 3, 1]
print reduce(reduce_func, lst)
With that small revision, the code produces 6 as you expected :-)

Your function calls fact() on both arguments. You are calculating ((1! * 3!)! * 1!). The workaround is to only call it on only the second argument, and pass reduce() an initial value of 1.

From the Python reduce documentation,
reduce(function, sequence) returns a single value constructed by calling the (binary) function on the first two items of the sequence, then on the result and the next item, and so on.
So, stepping through. It computes reduce_func of the first two elements, reduce_func(1, 3) = 1! * 3! = 6. Then, it computes reduce_func of the result and the next item: reduce_func(6, 1) = 6! * 1! = 720.
You missed that, when the result of the first reduce_func call is passed as input to the second, it's factorialized before the multiplication.

Ok, got it:
I need to map the numbers to their factorials first and then call reduce with multiply operator.
So, this would work:
lst_fact = map(fact, lst)
reduce(operator.mul, lst_fact)

You could also implement factorial using reduce.
def factorial(n):
return(reduce(lambda x,y:x*y,range(n+1)[1:]))

Beyond the trivial examples, here is one where I find reduce to be actually quite useful:
Imagine an iterable of ordered int values, often with some runs of contiguous values, and that we'd like to "summarize" it as a list of tuples representing ranges. (Note also that this iterable could be a generator of a very long sequence --another reason to use reduce and not some operation on an in-memory collection).
from functools import reduce
def rle(a, b):
if a and a[-1][1] == b:
return a[:-1] + [(a[-1][0], b + 1)]
return a + [(b, b + 1)]
reduce(rle, [0, 1, 2, 5, 8, 9], [])
# [(0, 3), (5, 6), (8, 10)]
Notice the use of a proper initial value ([] here) for reduce.
Corner cases handled as well:
reduce(rle, [], [])
# []
reduce(rle, [0], [])
# [(0, 1)]

Well, first of all, your reduce_func doesn't have the structure of a fold; it doesn't match your description of a fold (which is correct).
The structure of a fold is: def foldl(func, start, iter): return func(start, foldl(func, next(iter), iter)
Now, your fact function doesn't operate on two elements - it just calculates factorial.
So, in sum, you're not using a fold, and with that definition of factorial, you don't need to.
If you do want to play around with factorial, check out the y-combinator: http://mvanier.livejournal.com/2897.html
If you want to learn about folds, look at my answer to this question, which demonstrates its use to calculate cumulative fractions: creating cumulative percentage from a dictionary of data

Reduce executes the function in parameter#1 successively through the values provided by the iterator in parameter#2
print '-------------- Example: Reduce(x + y) --------------'
def add(x,y): return x+y
x = 5
y = 10
import functools
tot = functools.reduce(add, range(5, 10))
print 'reduce('+str(x)+','+str(y)+')=' ,tot
def myreduce(a,b):
tot = 0
for i in range(a,b):
tot = tot+i
print i,tot
print 'myreduce('+str(a)+','+str(b)+')=' ,tot
myreduce(x,y)
print '-------------- Example: Reduce(x * y) --------------'
def add(x,y): return x*y
x = 5
y = 10
import functools
tot = functools.reduce(add, range(5, 10))
print 'reduce('+str(x)+','+str(y)+')=' ,tot
def myreduce(a,b):
tot = 1
for i in range(a,b):
tot = tot * i
print i,tot
print 'myreduce('+str(a)+','+str(b)+')=' ,tot
myreduce(x,y)

Related

Optimizing a factorial function in python

So i have achieved this function with unpacking parameter(*x), but i want to make it display the result not return it , and i want a good optimization meaning i still need it to be a two lines function
1.def fac(*x):
2.return (fac(list(x)[0], list(x)[1] - 1)*list(x)[1]) if list(x)[1] > 0 else 1//here i need the one line to print the factorial
i tried achieving this by implementing lambda but i didn't know how to pass the *x parameter
Your factorial lambda is correct. I take it that you would like to calculate the factorials for a list say [1, 2, 3] and output the results, this is how you can achieve this.
fact = lambda x: x*fact(x-1) if x > 0 else 1
print(*[fact(i) for i in [1, 2, 3]])
Which will output: 1, 2, 6
Another option, if you have python 3.8 is to use a list comprehension with the new walrus operator (:=), this is a bit more tricky but will calculate and output all factorials up to n inclusive whilst still fitting in your required two lines.
fac, n = 1, 5
print(*[fac for i in range(1, n+1) if (fac := fac*i)])
Which will output: 1, 2, 6, 24, 120
The optimized factorial number is display by the function that i have created below.
def fact(n):
list_fact = []
if n > 1 and n not in list_fact:
list_fact.extend(list(range(1, n + 1)))
return reduce(lambda x, y: x * y, list_fact)
print(fact(9000)) # it will display output within microseconds.
Note:
while iteration i saved all previous values into a list, so that computation of each value is not going to happen each time.

How to plug f(k) in to a series?

Lets suppose you have f(x)=x^2 and you want a series from k=1 to n of f(k). How would you enter it in python and print the result?
I tried doing this for n=10
n=10
def f(x):
return x^2
sum = 0
for k in range(1, n):
Sum = sum + f(k)
I don't know how to print it let alone whether I did it correctly.
I know you can represent a sum of k^2 as n(n+1)(2n+1)/6 but I am just using x^2 as an example.
First off sum is a builtin function so you should not use it as a variable name as python will simply let you overwrite the variable in the spirit of "we are all consenting adults here". Example
type(sum)
# <type 'builtin_function_or_method'>
sum = 0
type(sum)
# <type 'int'>
Now to your question. One way is to first generate the sequence and use the aforementioned sum function as the following examples show.
Generate the sequence using a so-called list comprehension to evaluate the function for each element in a range:
n = 5
def f(x):
return x*x
print([ f(i) for i in range(n)])
# [0, 1, 4, 9, 16]
print(sum([ f(i) for i in range(n)]))
# 30
Use the builtin map function to apply a function, either defined using def or lambda, on a range:
n = 10
f = lambda x: x*x
print(map(f, range(n)))
# [0, 1, 4, 9, 16]
# one line equivalent : print(map(lambda x: x*x, range(n)))
print(sum(map(f, range(n))))
# 30
Alternatively, you can follow your train of thought and use a simple loop and a variable:
n = 5
f = lambda x: x*x
s = 0
for i in range(n):
s += f(i)
print(s)
# 30
Finally, if speed/efficiency is a factor that you have to account for you can have a look at the numpy library. See for instance one of my answers on applying a funcion over a range.

How can I make a combined power function faster?

I made this function pretty quickly to find the combined powers of everything in a list, but I'm pretty sure there is a way to make it faster. It returns a 2 item list, with the combined powers and the list formatted for export.
#The input for this function is a list of numbers (but string data type)
def find(x):
total = int(x[0])
for i in range (1,len(x)):
total = total ** int(x[i])
value = [total,'^'.join(x)]
return value
This will calculate the values faster than what you currently have:
import functools
import operator
l = [2, 3, 4]
functools.reduce(operator.pow, l)
## 4096
If you want to show the chain of values in the list too, as in the original post, you can define a function e.g. like this one:
def func(vals):
vals_string = '^'.join(str(val) for val in vals)
total = functools.reduce(operator.pow, vals)
return [total, vals_string]
Usage example:
l = [2, 3, 4, 5]
result = func(l)
print(result)
## [1152921504606846976, '2^3^4^5']
You want to avoid doing a stack of exponentiations, as those are expensive. Multiplication is cheaper, so you can save computing power by multiplying all the values to the right of the first one, and then raising the first to that power.
from functools import reduce
from operator import mul
from typing import List
def find(x: List[int]) -> int:
return x[0]**reduce(mul, x[1:], 1)
You could also do
def find(i: int, *updog: int) -> int:
return i ** reduce(mul, updog, 1)
and call it like find(*[2,1,2,3,4]) or find(2, 1, 2, 3).
Either way, this uses the function signature to provide a stricter guarantee that the initial value is set, rather than having undefined behavior if the list is empty.
You can use NumPy's power.accumulate() to have cumulative exponents. However, you will soon run into large numbers for a large input list. The [-1] here gives you the cumulative of all the elements
import numpy as np
def find(x):
total = np.power.accumulate(x)[-1]
value = [total,'^'.join(map(str, inp))]
return value
x = np.array([2,3,4,5])
print (find(x))
# [1152921504606846976, '2^3^4^5']
You can use functools.reduce():
from functools import reduce
import operator
reduce(operator.pow, list)
What reduce() does is it applies the first parameter to every element in the second.

recursive function in python but with strange return

I am trying to solve a primary equation with several variables. For example:11x+7y+3z=20. non-negative integer result only.
I use code below in python 3.5.1, but the result contains something like [...]. I wonder what is it?
The code I have is to test every variables from 0 to max [total value divided by corresponding variable]. Because the variables may be of a large number, I want to use recursion to solve it.
def equation (a,b,relist):
global total
if len(a)>1:
for i in range(b//a[0]+1):
corelist=relist.copy()
corelist+=[i]
testrest=equation(a[1:],b-a[0]*i,corelist)
if testrest:
total+=[testrest]
return total
else:
if b%a[0]==0:
relist+=[b//a[0]]
return relist
else:
return False
total=[]
re=equation([11,7,3],20,[])
print(re)
the result is
[[0, 2, 2], [...], [1, 0, 3], [...]]
change to a new one could get clean result, but I still need a global variable:
def equation (a,b,relist):
global total
if len(a)>1:
for i in range(b//a[0]+1):
corelist=relist.copy()
corelist+=[i]
equation(a[1:],b-a[0]*i,corelist)
return total
else:
if b%a[0]==0:
relist+=[b//a[0]]
total+=[relist]
return
else:
return
total=[]
print(equation([11,7,3],20,[]))
I see three layers of problems here.
1) There seems to be a misunderstanding about recursion.
2) There seems to be an underestimation of the complexity of the problem you are trying to solve (a modeling issue)
3) Your main question exposes some lacking skills in python itself.
I will address the questions in backward order given that your actual question is "the result contains something like [...]. I wonder what is it?"
"[]" in python designates a list.
For example:
var = [ 1, 2 ,3 ,4 ]
Creates a reference "var" to a list containing 4 integers of values 1, 2, 3 and 4 respectively.
var2 = [ "hello", ["foo", "bar"], "world" ]
var2 on the other hand is a reference to a composite list of 3 elements, a string, another list and a string. The 2nd element is a list of 2 strings.
So your results is a list of lists of integers (assuming the 2 lists with "..." are integers). If each sublists are of the same size, you could also think of it as a matrix. And the way the function is written, you could end up with a composite list of lists of integers, the value "False" (or the value "None" in the newest version)
Now to the modeling problem. The equation 11x + 7y + 3z = 20 is one equation with 3 unknowns. It is not clear at all to me what you want to acheive with this program, but unless you solve the equation by selecting 2 independent variables, you won't achieve much. It is not clear at all to me what is the relation between the program and the equation save for the list you provided as argument with the values 11, 7 and 3.
What I would do (assuming you are looking for triplets of values that solves the equation) is go for the equation: f(x,y) = (20/3) - (11/3)x - (7/3)y. Then the code I would rather write is:
def func_f(x, y):
return 20.0/3.0 - (11.0/3.0) * x - (7.0/3.0) * y
list_of_list_of_triplets = []
for (x, y) in zip(range(100),range(100)):
list_of_triplet = [x, y, func_f(x,y)]
list_of_list_of_triplets += [list_of_triplet] # or .append(list_of_triplet)
Be mindful that the number of solutions to this equation is infinite. You could think of it as a straight line in a rectangular prism if you bound the variables. If you wanted to represent the same line in an abstract number of dimensions, you could rewrite the above as:
def func_multi_f(nthc, const, coeffs, vars):
return const - sum([a*b/nth for a,b in zip(coeffs, vars)])
Where nthc is the coefficient of the Nth variable, const is an offset constant, coeffs is a list of coefficients and vars the values of the N-1 other variables. For example, we could re-write the func_f as:
def func_f(x,y):
return func_multi_f(3.0, 20.0, [11.0, 7.0], [x,y])
Now about recursion. A recursion is a formulation of a reducible input that can be called repetivitely as to achieve a final result. In pseudo code a recursive algorithm can be formulated as:
input = a reduced value or input items
if input has reached final state: return final value
operation = perform something on input and reduce it, combine with return value of this algorithm with reduced input.
For example, the fibonacci suite:
def fibonacci(val):
if val == 1:
return 1
return fibonacci(val - 1) + val
If you wanted to recusively add elements from a list:
def sum_recursive(list):
if len(list) == 1:
return list[0]
return sum_recursive(list[:-1]) + list[-1]
Hope it helps.
UPDATE
From comments and original question edits, it appears that we are rather looking for INTEGER solutions to the equation. Of non-negative values. That is quite different.
1) Step one find bounds: use the equation ax + by + cz <= 20 with a,b,c > 0 and x,y,z >= 0
2) Step two, simply do [(x, y, z) for x, y, z in zip(bounds_x, bounds_y, bounds_z) if x*11 + y*7 + z*3 - 20 == 0] and you will have a list of valid triplets.
in code:
def bounds(coeff, const):
return [val for val in range(const) if coeff * val <= const]
def combine_bounds(bounds_list):
# here you have to write your recusive function to build
# all possible combinations assuming N dimensions
def sols(coeffs, const):
bounds_lists = [bounds(a, const) for a in coeffs]
return [vals for vals in combine_bounds(bounds_lists) if sum([a*b for a,b in zip(coeff, vals)] - const == 0)
Here is a solution built from your second one, but without the global variable. Instead, each call passes back a list of solutions; the parent call appends each solution to the current element, making a new list to return.
def equation (a, b):
result = []
if len(a) > 1:
# For each valid value of the current coefficient,
# recur on the remainder of the list.
for i in range(b // a[0]+1):
soln = equation(a[1:], b-a[0]*i)
# prepend the current coefficient
# to each solution of the recursive call.
for item in soln:
result.append([i] + item)
else:
# Only one item left: is it a solution?
if b%a[0] == 0:
# Success: return a list of the one element
result = [[b // a[0]]]
else:
# Failure: return empty list
result = []
return result
print(equation([11, 7, 3], 20, []))

Subsequent Application of Python First Class Functions: Any Way Cleaner Than Nesting?

I'm working an example to help me learn how to use first-class functions in Python. In general, I'm satisfied with the solution I came up with, except for one line of code that screams "un-Pythonic" to me.
So the problem I'm working with is defined here. The puzzle seeks the single permutation (out of 720 possible) of six simple functions involving "2" that ultimately returns -3.
Here's my solution, which simply dumps every possible six-function permutation and its result.
def perform (fun, arg):
return fun(arg)
def a(n):
return n + 2
def d(n):
return n / 2.
def m(n):
return n * 2
def p(n):
return n ** 2
def r(n):
return n ** 0.5
def s(n):
return n - 2
if __name__ == "__main__":
from itertools import permutations
for i, perm in enumerate(permutations([a, d, m, p, r, s])):
try:
k = perform(perm[5], perform(perm[4], perform(perm[3], perform(perm[2], perform(perm[1], perform(perm[0], 0))))))
except ValueError:
k = float('nan')
print "%03d. %s: %8.8f" % (i + 1, ''.join([x.__name__ for x in perm]), k)
The line that doesn't seem right to me is the one with nested perform calls: k = perform(...perform(...(. What I need to do is apply the first function in the permutation tuple to 0, and then that function's result to the second function in the tuple, and so on through the permutation tuple until I come up with the ultimate result of applying the component functions.
Is there a cleaner way to successively apply the functions in perm to the corresponding results, starting with 0 as an argument? I've toyed with map and recursion, but I haven't been able to hit upon a solution any more elegant than the one above.
Why not simply:
x = init value
for f in funclist:
x = f(x)
or in a bit fancier way:
value = reduce(lambda x, f: f(x), funclist, init_value)

Categories

Resources