Related
I'm trying to use recursion to find the colosest numbers in a list.
I made a program that runs corectly without any errors but I want to know if there's a better way to right my program while still using recursion.
This is my program:
deff = -1
num1 = 0
num2 = 0
def closest(List):
global deff, num1, num2
deff1 = -1
if deff == -1:
deff = max(List) - min(List)
deff1 = max(List) - min(List)
if len(List) == 1:
deff = -1
return [num1, num2]
if abs(List[0] - List[1]) <= deff:
deff = abs(List[0] - List[1])
num1 = List[0]
num2 = List[1]
return closest(List[1:])
This is my Tester Program:
from recursion import *
allPassed = True
def closestMain():
global allPassed
testCases = [(1, [3, 7, 67, 68, 210, 215], [67, 68]),
(2, [3, 7, 67, 168, 210, 215], [3, 7]),
(3, [3, 47, 67, 168, 210, 215], [210, 215]),
(4, [3, 7], [3, 7]),
(5, [3, 3, 3, 3, 3, 3], [3, 3]),
(6, [1, 2, 3, 4, 5, 6], [5, 6]),
(7, [5, 10, 100, 105, 305, 310], [305, 310]),
(8, [5, 10, 15], [10, 15])]
for num, L, expected in testCases:
result = closest(L)
if result != expected:
print(f'Closest Test {num} Failed. Expected {expected} got {result}')
allPassed = False
def main():
closestMain()
if allPassed:
print('All tests passed')
main()
Again no errors, the program works fine, just trying to see if there's a better way to do it using recursion.
Assume the list is always sorted and assume the length is always greater or equal to 2
here is my attempt, this algorithm works for sorted lists:
def closest(list_, close=None):
if len(list_) < 2: return close
if not close: return closest(list_[1:],list_[:2])
a,b = list_[:2]
x,y = close
return closest(list_[1:], (a,b) if b-a <= y-x else (x,y))
decomposed recursion
closest could be a recursive program, but it's a setup for a complex and tightly-coupled program. The high-level way to express the solution is using min and combinations, two generic functions, each of which could be implemented using recursion -
from itertools import combinations
def closest(l):
return min(combinations(l, 2), key=lambda pair: abs(pair[0] - pair[1]))
tests = [
(1, [3, 7, 67, 68, 210, 215], (67, 68)),
(2, [3, 7, 67, 168, 210, 215], (3, 7)),
(3, [3, 47, 67, 168, 210, 215], (210, 215)),
(4, [3, 7], (3, 7)),
(5, [3, 3, 3, 3, 3, 3], (3, 3)),
(6, [1, 2, 3, 4, 5, 6], (5, 6)),
(7, [5, 10, 100, 105, 305, 310], (305, 310)),
(8, [5, 10, 15], (10, 15))
]
for num, lst, expected in tests:
print(f"#{num} {expected} {closest(lst)}")
no.
expected
actual
1
(67, 68)
(67, 68)
2
(3, 7)
(3, 7)
3
(210, 215)
(210, 215)
4
(3, 7)
(3, 7)
5
(3, 3)
(3, 3)
6
(5, 6)
(1, 2)
7
(305, 310)
(5, 10)
8
(10, 15)
(5, 10)
Notice the earliest closest numbers are returned, unlike the latest that are expected in your output. Using the built-in min function we cannot control which minimum is returned, however we can supply our own recursive implementations and get the output we expect -
def combinations(t, n):
if n <= 0:
yield ()
elif not t:
return
else:
for x in combinations(t[1:], n - 1):
yield (t[0], *x)
yield from combinations(t[1:], n)
def min(t, key = lambda x: x):
def loop(a):
try:
b = next(t)
return loop(b if key(b) < key(a) else a)
except StopIteration:
return a
return loop(next(t, None))
If we rerun the tests, we see the output is the same. So how can we control it?
getting the exact output
Because we wrote min we now have the power to change how it behaves. Instead of returning the earliest minimum value in the series, we can return the latest. This is easily accomplished by reordering the if..else -
def min(t, key = lambda x: x):
def loop(a):
try:
b = next(t)
return loop(a if key(a) < key(b) else b) # <--
except StopIteration:
return a
return loop(next(t, None))
Rerun the tests to see the updated output -
for num, lst, expected in tests:
print(f"#{num} {expected} {closest(lst)}")
no.
expected
actual
#1
(67, 68)
(67, 68)
#2
(3, 7)
(3, 7)
#3
(210, 215)
(210, 215)
#4
(3, 7)
(3, 7)
#5
(3, 3)
(3, 3)
#6
(5, 6)
(5, 6)
#7
(305, 310)
(305, 310)
#8
(10, 15)
(10, 15)
using iterables correctly
Writing min using recursion is a fun exercise, but there's a more ergonomic way to interact with python's iterables. The simple for..in loop is fast and doesn't risk overflowing the stack -
def min(t, key = lambda x: x):
a = next(t, None)
for b in t:
a = a if key(a) < key(b) else b
return a
Rerun the tests and verify the output is identical.
Simple one without global variables and without changing the closest(List) signature (Try it online!):
def closest(List):
if len(List) == 2:
return List * 1
a, b = List[:2]
c, d = closest(List[1:])
if b - a < d - c:
return [a, b]
else:
return [c, d]
I have several people making x choices ranked in order of preference among n possibilities. Each possibility can only be assigned once.
I would like to find all the solutions to the problem such that each person has a choice of a minimum level xmin.
For example, for x=3 and say n=20, and 10 peoples making choices :
g1 = (3, 10, 11) # g1 makes choices 3, 10 and 11 in order of preference
g2 = (10, 9, 5)
g3 = (10, 15, 3)
g4 = (5, 9, 14)
g5 = (10, 3, 7)
...
g10 = (4, 19, 2)
With Python, how to write the problem to find the solutions so that all people have a choice with at least level 2 (xmin=2) assigned ? then level 3 (xmin=3) if there is no solution for xmin=2 ?
I think there is something to do with itertools, but I can't get a clear idea about this problem.
EDIT: Thinking a little bit more about the problem, I have something like this coming to me:
import itertools
xmin = 2
groups = [g1, g2, g3, g4, g5]
sample = [g[:xmin] for g in groups]
[seq for seq in itertools.product(*sample) if len(seq) == len(set(seq))]
I wasn't awake, the answer is simple in fact!
Finally, itertools simplifies life as always :)
With this dataset :
gr1 = (3, 10, 11)
gr2 = (10, 9, 5)
gr3 = (10, 15, 3)
gr4 = (5, 9, 14)
gr5 = (10, 22, 7)
gr6 = (24, 9, 17)
gr7 = (10, 17, 3)
gr8 = (14, 21, 2)
gr9 = (14, 12, 21)
gr10 = (9, 1, 22)
gr11 = (19, 2, 8)
gr12 = (3, 8, 24)
gr13 = (9, 5, 13)
gr14 = (9, 1, 12)
Here is a solution to the problem:
import itertools
import numpy as np
xmin = 3
groups = [gr1, gr2, gr3, gr4, gr5, gr6, gr7, gr8, gr9, gr10, gr11, gr12, gr13, gr14]
sample = [g[:xmin] for g in groups]
assign = [seq for seq in itertools.product(*sample) if len(seq) == len(set(seq))]
choices = [[sample[i].index(assign[j][i]) for i in range(len(sample))] for j in range(len(assign))]
print(f'{len(choices)} possibilities found')
pertinence = [sum(choices[i]) for i in range(len(choices))]
best = np.array(assign)[np.array(pertinence) == min(pertinence)]
print(f'{len(best)} best choices:')
print(best)
If someone has a more elegant solution...
I'd like to identify groups of consecutive numbers in a list, so that:
myfunc([2, 3, 4, 5, 12, 13, 14, 15, 16, 17, 20])
Returns:
[(2,5), (12,17), 20]
And was wondering what the best way to do this was (particularly if there's something inbuilt into Python).
Edit: Note I originally forgot to mention that individual numbers should be returned as individual numbers, not ranges.
EDIT 2: To answer the OP new requirement
ranges = []
for key, group in groupby(enumerate(data), lambda (index, item): index - item):
group = map(itemgetter(1), group)
if len(group) > 1:
ranges.append(xrange(group[0], group[-1]))
else:
ranges.append(group[0])
Output:
[xrange(2, 5), xrange(12, 17), 20]
You can replace xrange with range or any other custom class.
Python docs have a very neat recipe for this:
from operator import itemgetter
from itertools import groupby
data = [2, 3, 4, 5, 12, 13, 14, 15, 16, 17]
for k, g in groupby(enumerate(data), lambda (i,x):i-x):
print(map(itemgetter(1), g))
Output:
[2, 3, 4, 5]
[12, 13, 14, 15, 16, 17]
If you want to get the exact same output, you can do this:
ranges = []
for k, g in groupby(enumerate(data), lambda (i,x):i-x):
group = map(itemgetter(1), g)
ranges.append((group[0], group[-1]))
output:
[(2, 5), (12, 17)]
EDIT: The example is already explained in the documentation but maybe I should explain it more:
The key to the solution is
differencing with a range so that
consecutive numbers all appear in same
group.
If the data was: [2, 3, 4, 5, 12, 13, 14, 15, 16, 17]
Then groupby(enumerate(data), lambda (i,x):i-x) is equivalent of the following:
groupby(
[(0, 2), (1, 3), (2, 4), (3, 5), (4, 12),
(5, 13), (6, 14), (7, 15), (8, 16), (9, 17)],
lambda (i,x):i-x
)
The lambda function subtracts the element index from the element value. So when you apply the lambda on each item. You'll get the following keys for groupby:
[-2, -2, -2, -2, -8, -8, -8, -8, -8, -8]
groupby groups elements by equal key value, so the first 4 elements will be grouped together and so forth.
I hope this makes it more readable.
python 3 version may be helpful for beginners
import the libraries required first
from itertools import groupby
from operator import itemgetter
ranges =[]
for k,g in groupby(enumerate(data),lambda x:x[0]-x[1]):
group = (map(itemgetter(1),g))
group = list(map(int,group))
ranges.append((group[0],group[-1]))
more_itertools.consecutive_groups was added in version 4.0.
Demo
import more_itertools as mit
iterable = [2, 3, 4, 5, 12, 13, 14, 15, 16, 17, 20]
[list(group) for group in mit.consecutive_groups(iterable)]
# [[2, 3, 4, 5], [12, 13, 14, 15, 16, 17], [20]]
Code
Applying this tool, we make a generator function that finds ranges of consecutive numbers.
def find_ranges(iterable):
"""Yield range of consecutive numbers."""
for group in mit.consecutive_groups(iterable):
group = list(group)
if len(group) == 1:
yield group[0]
else:
yield group[0], group[-1]
iterable = [2, 3, 4, 5, 12, 13, 14, 15, 16, 17, 20]
list(find_ranges(iterable))
# [(2, 5), (12, 17), 20]
The source implementation emulates a classic recipe (as demonstrated by #Nadia Alramli).
Note: more_itertools is a third-party package installable via pip install more_itertools.
The "naive" solution which I find somewhat readable atleast.
x = [2, 3, 4, 5, 12, 13, 14, 15, 16, 17, 22, 25, 26, 28, 51, 52, 57]
def group(L):
first = last = L[0]
for n in L[1:]:
if n - 1 == last: # Part of the group, bump the end
last = n
else: # Not part of the group, yield current group and start a new
yield first, last
first = last = n
yield first, last # Yield the last group
>>>print list(group(x))
[(2, 5), (12, 17), (22, 22), (25, 26), (28, 28), (51, 52), (57, 57)]
Assuming your list is sorted:
>>> from itertools import groupby
>>> def ranges(lst):
pos = (j - i for i, j in enumerate(lst))
t = 0
for i, els in groupby(pos):
l = len(list(els))
el = lst[t]
t += l
yield range(el, el+l)
>>> lst = [2, 3, 4, 5, 12, 13, 14, 15, 16, 17]
>>> list(ranges(lst))
[range(2, 6), range(12, 18)]
Here it is something that should work, without any import needed:
def myfunc(lst):
ret = []
a = b = lst[0] # a and b are range's bounds
for el in lst[1:]:
if el == b+1:
b = el # range grows
else: # range ended
ret.append(a if a==b else (a,b)) # is a single or a range?
a = b = el # let's start again with a single
ret.append(a if a==b else (a,b)) # corner case for last single/range
return ret
Please note that the code using groupby doesn't work as given in Python 3 so use this.
for k, g in groupby(enumerate(data), lambda x:x[0]-x[1]):
group = list(map(itemgetter(1), g))
ranges.append((group[0], group[-1]))
This doesn't use a standard function - it just iiterates over the input, but it should work:
def myfunc(l):
r = []
p = q = None
for x in l + [-1]:
if x - 1 == q:
q += 1
else:
if p:
if q > p:
r.append('%s-%s' % (p, q))
else:
r.append(str(p))
p = q = x
return '(%s)' % ', '.join(r)
Note that it requires that the input contains only positive numbers in ascending order. You should validate the input, but this code is omitted for clarity.
import numpy as np
myarray = [2, 3, 4, 5, 12, 13, 14, 15, 16, 17, 20]
sequences = np.split(myarray, np.array(np.where(np.diff(myarray) > 1)[0]) + 1)
l = []
for s in sequences:
if len(s) > 1:
l.append((np.min(s), np.max(s)))
else:
l.append(s[0])
print(l)
Output:
[(2, 5), (12, 17), 20]
I think this way is simpler than any of the answers I've seen here (Edit: fixed based on comment from Pleastry):
data = [2, 3, 4, 5, 12, 13, 14, 15, 16, 17, 20]
starts = [x for x in data if x-1 not in data and x+1 in data]
ends = [x for x in data if x-1 in data and x+1 not in data and x not in starts]
singles = [x for x in data if x-1 not in data and x+1 not in data]
list(zip(starts, ends)) + singles
Output:
[(2, 5), (12, 17), 20]
Edited:
As #dawg notes, this is O(n**2). One option to improve performance would be to convert the original list to a set (and also the starts list to a set) i.e.
data = [2, 3, 4, 5, 12, 13, 14, 15, 16, 17, 20]
data_as_set = set(data)
starts = [x for x in data_as_set if x-1 not in data_as_set and x+1 in data_as_set]
startset = set(starts)
ends = [x for x in data_as_set if x-1 in data_as_set and x+1 not in data_as_set and x not in startset]
singles = [x for x in data_as_set if x-1 not in data_as_set and x+1 not in data_as_set]
print(list(zip(starts, ends)) + singles)
Using groupby and count from itertools gives us a short solution. The idea is that, in an increasing sequence, the difference between the index and the value will remain the same.
In order to keep track of the index, we can use an itertools.count, which makes the code cleaner as using enumerate:
from itertools import groupby, count
def intervals(data):
out = []
counter = count()
for key, group in groupby(data, key = lambda x: x-next(counter)):
block = list(group)
out.append([block[0], block[-1]])
return out
Some sample output:
print(intervals([0, 1, 3, 4, 6]))
# [[0, 1], [3, 4], [6, 6]]
print(intervals([2, 3, 4, 5]))
# [[2, 5]]
This is my method in which I tried to prioritize readability. Note that it returns a tuple of the same values if there is only one value in a group. That can be fixed easily in the second snippet I'll post.
def group(values):
"""return the first and last value of each continuous set in a list of sorted values"""
values = sorted(values)
first = last = values[0]
for index in values[1:]:
if index - last > 1: # triggered if in a new group
yield first, last
first = index # update first only if in a new group
last = index # update last on every iteration
yield first, last # this is needed to yield the last set of numbers
Here is the result of a test:
values = [0, 5, 6, 7, 12, 13, 21, 22, 23, 24, 25, 26, 30, 44, 45, 50]
result = list(group(values))
print(result)
result = [(0, 0), (5, 7), (12, 13), (21, 26), (30, 30), (44, 45), (50, 50)]
If you want to return only a single value in the case of a single value in a group, just add a conditional check to the yields:
def group(values):
"""return the first and last value of each continuous set in a list of sorted values"""
values = sorted(values)
first = last = values[0]
for index in values[1:]:
if index - last > 1: # triggered if in a new group
if first == last:
yield first
else:
yield first, last
first = index # update first only if in a new group
last = index # update last on every iteration
if first == last:
yield first
else:
yield first, last
result = [0, (5, 7), (12, 13), (21, 26), 30, (44, 45), 50]
Here's the answer I came up with. I'm writing the code for other people to understand, so I'm fairly verbose with variable names and comments.
First a quick helper function:
def getpreviousitem(mylist,myitem):
'''Given a list and an item, return previous item in list'''
for position, item in enumerate(mylist):
if item == myitem:
# First item has no previous item
if position == 0:
return None
# Return previous item
return mylist[position-1]
And then the actual code:
def getranges(cpulist):
'''Given a sorted list of numbers, return a list of ranges'''
rangelist = []
inrange = False
for item in cpulist:
previousitem = getpreviousitem(cpulist,item)
if previousitem == item - 1:
# We're in a range
if inrange == True:
# It's an existing range - change the end to the current item
newrange[1] = item
else:
# We've found a new range.
newrange = [item-1,item]
# Update to show we are now in a range
inrange = True
else:
# We were in a range but now it just ended
if inrange == True:
# Save the old range
rangelist.append(newrange)
# Update to show we're no longer in a range
inrange = False
# Add the final range found to our list
if inrange == True:
rangelist.append(newrange)
return rangelist
Example run:
getranges([2, 3, 4, 5, 12, 13, 14, 15, 16, 17])
returns:
[[2, 5], [12, 17]]
Using numpy + comprehension lists:
With numpy diff function, consequent input vector entries that their difference is not equal to one can be identified. The start and end of the input vector need to be considered.
import numpy as np
data = np.array([2, 3, 4, 5, 12, 13, 14, 15, 16, 17, 20])
d = [i for i, df in enumerate(np.diff(data)) if df!= 1]
d = np.hstack([-1, d, len(data)-1]) # add first and last elements
d = np.vstack([d[:-1]+1, d[1:]]).T
print(data[d])
Output:
[[ 2 5]
[12 17]
[20 20]]
Note: The request that individual numbers should be treated differently, (returned as individual, not ranges) was omitted. This can be reached by further post-processing the results. Usually this will make things more complex without gaining any benefit.
One-liner in Python 2.7 if interested:
x = [2, 3, 6, 7, 8, 14, 15, 19, 20, 21]
d = iter(x[:1] + sum(([i1, i2] for i1, i2 in zip(x, x[1:] + x[:1]) if i2 != i1+1), []))
print zip(d, d)
>>> [(2, 3), (6, 8), (14, 15), (19, 21)]
A short solution that works without additional imports. It accepts any iterable, sorts unsorted inputs, and removes duplicate items:
def ranges(nums):
nums = sorted(set(nums))
gaps = [[s, e] for s, e in zip(nums, nums[1:]) if s+1 < e]
edges = iter(nums[:1] + sum(gaps, []) + nums[-1:])
return list(zip(edges, edges))
Example:
>>> ranges([2, 3, 4, 7, 8, 9, 15])
[(2, 4), (7, 9), (15, 15)]
>>> ranges([-1, 0, 1, 2, 3, 12, 13, 15, 100])
[(-1, 3), (12, 13), (15, 15), (100, 100)]
>>> ranges(range(100))
[(0, 99)]
>>> ranges([0])
[(0, 0)]
>>> ranges([])
[]
This is the same as #dansalmo's solution which I found amazing, albeit a bit hard to read and apply (as it's not given as a function).
Note that it could easily be modified to spit out "traditional" open ranges [start, end), by e.g. altering the return statement:
return [(s, e+1) for s, e in zip(edges, edges)]
I copied this answer over from another question that was marked as a duplicate of this one with the intention to make it easier findable (after I just now searched again for this topic, finding only the question here at first and not being satisfied with the answers given).
The versions by Mark Byers, Andrea Ambu, SilentGhost, Nadia Alramli, and truppo are simple and fast. The 'truppo' version encouraged me to write a version that retains the same nimble behavior while handling step sizes other than 1 (and lists as singletons elements that don't extend more than 1 step with a given step size). It is given here.
>>> list(ranges([1,2,3,4,3,2,1,3,5,7,11,1,2,3]))
[(1, 4, 1), (3, 1, -1), (3, 7, 2), 11, (1, 3, 1)]
Not the best approach , but here is my 2 cents
def getConsecutiveValues2(arr):
x = ""
final = []
end = 0
start = 0
for i in range(1,len(arr)) :
if arr[i] - arr[i-1] == 1 :
end = i
else :
print(start,end)
final.append(arr[start:end+1])
start = i
if i == len(arr) - 1 :
final.append(arr[start:end+1])
return final
x = [1,2,3,5,6,8,9,10,11,12]
print(getConsecutiveValues2(x))
>> [[1, 2, 3], [5, 6], [8, 9, 10, 11]]
This implementation works for regular or irregular steps
I needed to achieve the same thing but with the slight difference where steps can be irregular. this is my implementation
def ranges(l):
if not len(l):
return range(0,0)
elif len(l)==1:
return range(l[0],l[0]+1)
# get steps
sl = sorted(l)
steps = [i-j for i,j in zip(sl[1:],sl[:-1])]
# get unique steps indexes range
groups = [[0,0,steps[0]],]
for i,s in enumerate(steps):
if s==groups[-1][-1]:
groups[-1][1] = i+1
else:
groups.append( [i+1,i+1,s] )
g2 = groups[-2]
if g2[0]==g2[1]:
if sl[i+1]-sl[i]==s:
_=groups.pop(-2)
groups[-1][0] = i
# create list of ranges
return [range(sl[i],sl[j]+s,s) if s!=0 else [sl[i]]*(j+1-i) for i,j,s in groups]
Here's an example
from timeit import timeit
# for regular ranges
l = list(range(1000000))
ranges(l)
>>> [range(0, 1000000)]
l = list(range(10)) + list(range(20,25)) + [1,2,3]
ranges(l)
>>> [range(0, 2), range(1, 3), range(2, 4), range(3, 10), range(20, 25)]
sorted(l);[list(i) for i in ranges(l)]
>>> [0, 1, 1, 2, 2, 3, 3, 4, 5, 6, 7, 8, 9, 20, 21, 22, 23, 24]
>>> [[0, 1], [1, 2], [2, 3], [3, 4, 5, 6, 7, 8, 9], [20, 21, 22, 23, 24]]
# for irregular steps list
l = [1, 3, 5, 7, 10, 11, 12, 100, 200, 300, 400, 60, 99, 4000,4001]
ranges(l)
>>> [range(1, 9, 2), range(10, 13), range(60, 138, 39), range(100, 500, 100), range(4000, 4002)]
## Speed test
timeit("ranges(l)","from __main__ import ranges,l", number=1000)/1000
>>> 9.303160999934334e-06
Yet another solution if you expect your input to be a set:
def group_years(years):
consecutive_years = []
for year in years:
close = {y for y in years if abs(y - year) == 1}
for group in consecutive_years:
if len(close.intersection(group)):
group |= close
break
else:
consecutive_years.append({year, *close})
return consecutive_years
Example:
group_years({2016, 2017, 2019, 2020, 2022})
Out[54]: [{2016, 2017}, {2019, 2020}, {2022}]
I'd like to identify groups of consecutive numbers in a list, so that:
myfunc([2, 3, 4, 5, 12, 13, 14, 15, 16, 17, 20])
Returns:
[(2,5), (12,17), 20]
And was wondering what the best way to do this was (particularly if there's something inbuilt into Python).
Edit: Note I originally forgot to mention that individual numbers should be returned as individual numbers, not ranges.
EDIT 2: To answer the OP new requirement
ranges = []
for key, group in groupby(enumerate(data), lambda (index, item): index - item):
group = map(itemgetter(1), group)
if len(group) > 1:
ranges.append(xrange(group[0], group[-1]))
else:
ranges.append(group[0])
Output:
[xrange(2, 5), xrange(12, 17), 20]
You can replace xrange with range or any other custom class.
Python docs have a very neat recipe for this:
from operator import itemgetter
from itertools import groupby
data = [2, 3, 4, 5, 12, 13, 14, 15, 16, 17]
for k, g in groupby(enumerate(data), lambda (i,x):i-x):
print(map(itemgetter(1), g))
Output:
[2, 3, 4, 5]
[12, 13, 14, 15, 16, 17]
If you want to get the exact same output, you can do this:
ranges = []
for k, g in groupby(enumerate(data), lambda (i,x):i-x):
group = map(itemgetter(1), g)
ranges.append((group[0], group[-1]))
output:
[(2, 5), (12, 17)]
EDIT: The example is already explained in the documentation but maybe I should explain it more:
The key to the solution is
differencing with a range so that
consecutive numbers all appear in same
group.
If the data was: [2, 3, 4, 5, 12, 13, 14, 15, 16, 17]
Then groupby(enumerate(data), lambda (i,x):i-x) is equivalent of the following:
groupby(
[(0, 2), (1, 3), (2, 4), (3, 5), (4, 12),
(5, 13), (6, 14), (7, 15), (8, 16), (9, 17)],
lambda (i,x):i-x
)
The lambda function subtracts the element index from the element value. So when you apply the lambda on each item. You'll get the following keys for groupby:
[-2, -2, -2, -2, -8, -8, -8, -8, -8, -8]
groupby groups elements by equal key value, so the first 4 elements will be grouped together and so forth.
I hope this makes it more readable.
python 3 version may be helpful for beginners
import the libraries required first
from itertools import groupby
from operator import itemgetter
ranges =[]
for k,g in groupby(enumerate(data),lambda x:x[0]-x[1]):
group = (map(itemgetter(1),g))
group = list(map(int,group))
ranges.append((group[0],group[-1]))
more_itertools.consecutive_groups was added in version 4.0.
Demo
import more_itertools as mit
iterable = [2, 3, 4, 5, 12, 13, 14, 15, 16, 17, 20]
[list(group) for group in mit.consecutive_groups(iterable)]
# [[2, 3, 4, 5], [12, 13, 14, 15, 16, 17], [20]]
Code
Applying this tool, we make a generator function that finds ranges of consecutive numbers.
def find_ranges(iterable):
"""Yield range of consecutive numbers."""
for group in mit.consecutive_groups(iterable):
group = list(group)
if len(group) == 1:
yield group[0]
else:
yield group[0], group[-1]
iterable = [2, 3, 4, 5, 12, 13, 14, 15, 16, 17, 20]
list(find_ranges(iterable))
# [(2, 5), (12, 17), 20]
The source implementation emulates a classic recipe (as demonstrated by #Nadia Alramli).
Note: more_itertools is a third-party package installable via pip install more_itertools.
The "naive" solution which I find somewhat readable atleast.
x = [2, 3, 4, 5, 12, 13, 14, 15, 16, 17, 22, 25, 26, 28, 51, 52, 57]
def group(L):
first = last = L[0]
for n in L[1:]:
if n - 1 == last: # Part of the group, bump the end
last = n
else: # Not part of the group, yield current group and start a new
yield first, last
first = last = n
yield first, last # Yield the last group
>>>print list(group(x))
[(2, 5), (12, 17), (22, 22), (25, 26), (28, 28), (51, 52), (57, 57)]
Assuming your list is sorted:
>>> from itertools import groupby
>>> def ranges(lst):
pos = (j - i for i, j in enumerate(lst))
t = 0
for i, els in groupby(pos):
l = len(list(els))
el = lst[t]
t += l
yield range(el, el+l)
>>> lst = [2, 3, 4, 5, 12, 13, 14, 15, 16, 17]
>>> list(ranges(lst))
[range(2, 6), range(12, 18)]
Here it is something that should work, without any import needed:
def myfunc(lst):
ret = []
a = b = lst[0] # a and b are range's bounds
for el in lst[1:]:
if el == b+1:
b = el # range grows
else: # range ended
ret.append(a if a==b else (a,b)) # is a single or a range?
a = b = el # let's start again with a single
ret.append(a if a==b else (a,b)) # corner case for last single/range
return ret
Please note that the code using groupby doesn't work as given in Python 3 so use this.
for k, g in groupby(enumerate(data), lambda x:x[0]-x[1]):
group = list(map(itemgetter(1), g))
ranges.append((group[0], group[-1]))
This doesn't use a standard function - it just iiterates over the input, but it should work:
def myfunc(l):
r = []
p = q = None
for x in l + [-1]:
if x - 1 == q:
q += 1
else:
if p:
if q > p:
r.append('%s-%s' % (p, q))
else:
r.append(str(p))
p = q = x
return '(%s)' % ', '.join(r)
Note that it requires that the input contains only positive numbers in ascending order. You should validate the input, but this code is omitted for clarity.
import numpy as np
myarray = [2, 3, 4, 5, 12, 13, 14, 15, 16, 17, 20]
sequences = np.split(myarray, np.array(np.where(np.diff(myarray) > 1)[0]) + 1)
l = []
for s in sequences:
if len(s) > 1:
l.append((np.min(s), np.max(s)))
else:
l.append(s[0])
print(l)
Output:
[(2, 5), (12, 17), 20]
I think this way is simpler than any of the answers I've seen here (Edit: fixed based on comment from Pleastry):
data = [2, 3, 4, 5, 12, 13, 14, 15, 16, 17, 20]
starts = [x for x in data if x-1 not in data and x+1 in data]
ends = [x for x in data if x-1 in data and x+1 not in data and x not in starts]
singles = [x for x in data if x-1 not in data and x+1 not in data]
list(zip(starts, ends)) + singles
Output:
[(2, 5), (12, 17), 20]
Edited:
As #dawg notes, this is O(n**2). One option to improve performance would be to convert the original list to a set (and also the starts list to a set) i.e.
data = [2, 3, 4, 5, 12, 13, 14, 15, 16, 17, 20]
data_as_set = set(data)
starts = [x for x in data_as_set if x-1 not in data_as_set and x+1 in data_as_set]
startset = set(starts)
ends = [x for x in data_as_set if x-1 in data_as_set and x+1 not in data_as_set and x not in startset]
singles = [x for x in data_as_set if x-1 not in data_as_set and x+1 not in data_as_set]
print(list(zip(starts, ends)) + singles)
Using groupby and count from itertools gives us a short solution. The idea is that, in an increasing sequence, the difference between the index and the value will remain the same.
In order to keep track of the index, we can use an itertools.count, which makes the code cleaner as using enumerate:
from itertools import groupby, count
def intervals(data):
out = []
counter = count()
for key, group in groupby(data, key = lambda x: x-next(counter)):
block = list(group)
out.append([block[0], block[-1]])
return out
Some sample output:
print(intervals([0, 1, 3, 4, 6]))
# [[0, 1], [3, 4], [6, 6]]
print(intervals([2, 3, 4, 5]))
# [[2, 5]]
This is my method in which I tried to prioritize readability. Note that it returns a tuple of the same values if there is only one value in a group. That can be fixed easily in the second snippet I'll post.
def group(values):
"""return the first and last value of each continuous set in a list of sorted values"""
values = sorted(values)
first = last = values[0]
for index in values[1:]:
if index - last > 1: # triggered if in a new group
yield first, last
first = index # update first only if in a new group
last = index # update last on every iteration
yield first, last # this is needed to yield the last set of numbers
Here is the result of a test:
values = [0, 5, 6, 7, 12, 13, 21, 22, 23, 24, 25, 26, 30, 44, 45, 50]
result = list(group(values))
print(result)
result = [(0, 0), (5, 7), (12, 13), (21, 26), (30, 30), (44, 45), (50, 50)]
If you want to return only a single value in the case of a single value in a group, just add a conditional check to the yields:
def group(values):
"""return the first and last value of each continuous set in a list of sorted values"""
values = sorted(values)
first = last = values[0]
for index in values[1:]:
if index - last > 1: # triggered if in a new group
if first == last:
yield first
else:
yield first, last
first = index # update first only if in a new group
last = index # update last on every iteration
if first == last:
yield first
else:
yield first, last
result = [0, (5, 7), (12, 13), (21, 26), 30, (44, 45), 50]
Here's the answer I came up with. I'm writing the code for other people to understand, so I'm fairly verbose with variable names and comments.
First a quick helper function:
def getpreviousitem(mylist,myitem):
'''Given a list and an item, return previous item in list'''
for position, item in enumerate(mylist):
if item == myitem:
# First item has no previous item
if position == 0:
return None
# Return previous item
return mylist[position-1]
And then the actual code:
def getranges(cpulist):
'''Given a sorted list of numbers, return a list of ranges'''
rangelist = []
inrange = False
for item in cpulist:
previousitem = getpreviousitem(cpulist,item)
if previousitem == item - 1:
# We're in a range
if inrange == True:
# It's an existing range - change the end to the current item
newrange[1] = item
else:
# We've found a new range.
newrange = [item-1,item]
# Update to show we are now in a range
inrange = True
else:
# We were in a range but now it just ended
if inrange == True:
# Save the old range
rangelist.append(newrange)
# Update to show we're no longer in a range
inrange = False
# Add the final range found to our list
if inrange == True:
rangelist.append(newrange)
return rangelist
Example run:
getranges([2, 3, 4, 5, 12, 13, 14, 15, 16, 17])
returns:
[[2, 5], [12, 17]]
Using numpy + comprehension lists:
With numpy diff function, consequent input vector entries that their difference is not equal to one can be identified. The start and end of the input vector need to be considered.
import numpy as np
data = np.array([2, 3, 4, 5, 12, 13, 14, 15, 16, 17, 20])
d = [i for i, df in enumerate(np.diff(data)) if df!= 1]
d = np.hstack([-1, d, len(data)-1]) # add first and last elements
d = np.vstack([d[:-1]+1, d[1:]]).T
print(data[d])
Output:
[[ 2 5]
[12 17]
[20 20]]
Note: The request that individual numbers should be treated differently, (returned as individual, not ranges) was omitted. This can be reached by further post-processing the results. Usually this will make things more complex without gaining any benefit.
One-liner in Python 2.7 if interested:
x = [2, 3, 6, 7, 8, 14, 15, 19, 20, 21]
d = iter(x[:1] + sum(([i1, i2] for i1, i2 in zip(x, x[1:] + x[:1]) if i2 != i1+1), []))
print zip(d, d)
>>> [(2, 3), (6, 8), (14, 15), (19, 21)]
A short solution that works without additional imports. It accepts any iterable, sorts unsorted inputs, and removes duplicate items:
def ranges(nums):
nums = sorted(set(nums))
gaps = [[s, e] for s, e in zip(nums, nums[1:]) if s+1 < e]
edges = iter(nums[:1] + sum(gaps, []) + nums[-1:])
return list(zip(edges, edges))
Example:
>>> ranges([2, 3, 4, 7, 8, 9, 15])
[(2, 4), (7, 9), (15, 15)]
>>> ranges([-1, 0, 1, 2, 3, 12, 13, 15, 100])
[(-1, 3), (12, 13), (15, 15), (100, 100)]
>>> ranges(range(100))
[(0, 99)]
>>> ranges([0])
[(0, 0)]
>>> ranges([])
[]
This is the same as #dansalmo's solution which I found amazing, albeit a bit hard to read and apply (as it's not given as a function).
Note that it could easily be modified to spit out "traditional" open ranges [start, end), by e.g. altering the return statement:
return [(s, e+1) for s, e in zip(edges, edges)]
I copied this answer over from another question that was marked as a duplicate of this one with the intention to make it easier findable (after I just now searched again for this topic, finding only the question here at first and not being satisfied with the answers given).
The versions by Mark Byers, Andrea Ambu, SilentGhost, Nadia Alramli, and truppo are simple and fast. The 'truppo' version encouraged me to write a version that retains the same nimble behavior while handling step sizes other than 1 (and lists as singletons elements that don't extend more than 1 step with a given step size). It is given here.
>>> list(ranges([1,2,3,4,3,2,1,3,5,7,11,1,2,3]))
[(1, 4, 1), (3, 1, -1), (3, 7, 2), 11, (1, 3, 1)]
Not the best approach , but here is my 2 cents
def getConsecutiveValues2(arr):
x = ""
final = []
end = 0
start = 0
for i in range(1,len(arr)) :
if arr[i] - arr[i-1] == 1 :
end = i
else :
print(start,end)
final.append(arr[start:end+1])
start = i
if i == len(arr) - 1 :
final.append(arr[start:end+1])
return final
x = [1,2,3,5,6,8,9,10,11,12]
print(getConsecutiveValues2(x))
>> [[1, 2, 3], [5, 6], [8, 9, 10, 11]]
This implementation works for regular or irregular steps
I needed to achieve the same thing but with the slight difference where steps can be irregular. this is my implementation
def ranges(l):
if not len(l):
return range(0,0)
elif len(l)==1:
return range(l[0],l[0]+1)
# get steps
sl = sorted(l)
steps = [i-j for i,j in zip(sl[1:],sl[:-1])]
# get unique steps indexes range
groups = [[0,0,steps[0]],]
for i,s in enumerate(steps):
if s==groups[-1][-1]:
groups[-1][1] = i+1
else:
groups.append( [i+1,i+1,s] )
g2 = groups[-2]
if g2[0]==g2[1]:
if sl[i+1]-sl[i]==s:
_=groups.pop(-2)
groups[-1][0] = i
# create list of ranges
return [range(sl[i],sl[j]+s,s) if s!=0 else [sl[i]]*(j+1-i) for i,j,s in groups]
Here's an example
from timeit import timeit
# for regular ranges
l = list(range(1000000))
ranges(l)
>>> [range(0, 1000000)]
l = list(range(10)) + list(range(20,25)) + [1,2,3]
ranges(l)
>>> [range(0, 2), range(1, 3), range(2, 4), range(3, 10), range(20, 25)]
sorted(l);[list(i) for i in ranges(l)]
>>> [0, 1, 1, 2, 2, 3, 3, 4, 5, 6, 7, 8, 9, 20, 21, 22, 23, 24]
>>> [[0, 1], [1, 2], [2, 3], [3, 4, 5, 6, 7, 8, 9], [20, 21, 22, 23, 24]]
# for irregular steps list
l = [1, 3, 5, 7, 10, 11, 12, 100, 200, 300, 400, 60, 99, 4000,4001]
ranges(l)
>>> [range(1, 9, 2), range(10, 13), range(60, 138, 39), range(100, 500, 100), range(4000, 4002)]
## Speed test
timeit("ranges(l)","from __main__ import ranges,l", number=1000)/1000
>>> 9.303160999934334e-06
Yet another solution if you expect your input to be a set:
def group_years(years):
consecutive_years = []
for year in years:
close = {y for y in years if abs(y - year) == 1}
for group in consecutive_years:
if len(close.intersection(group)):
group |= close
break
else:
consecutive_years.append({year, *close})
return consecutive_years
Example:
group_years({2016, 2017, 2019, 2020, 2022})
Out[54]: [{2016, 2017}, {2019, 2020}, {2022}]
So here's the issue, I have a 2-d list of characters 'T' and 'F', and given coordinates I need to get all of its neighbors. I have this:
from itertools import product, starmap
x, y = (5, 5)
cells = starmap(lambda a, b: (x + a, y + b), product((0, -1, +1), (0, -1, +1)))
from determining neighbors of cell two dimensional list But it will only give me a list of coordinantes, so i still fetch the values afterwords. I'd like the search and retrieval done in one step, so findNeighbors(5,5) would return F,T,F,F,... instead of (5, 4), (5, 6), (4, 5), (4, 4)... Is there a fast way of doing this? The solutin can include a structure other than a list to hold the initial information
The following should work, with just a minor adaptation to the current code:
from itertools import product, starmap, islice
def findNeighbors(grid, x, y):
xi = (0, -1, 1) if 0 < x < len(grid) - 1 else ((0, -1) if x > 0 else (0, 1))
yi = (0, -1, 1) if 0 < y < len(grid[0]) - 1 else ((0, -1) if y > 0 else (0, 1))
return islice(starmap((lambda a, b: grid[x + a][y + b]), product(xi, yi)), 1, None)
For example:
>>> grid = [[ 0, 1, 2, 3],
... [ 4, 5, 6, 7],
... [ 8, 9, 10, 11],
... [12, 13, 14, 15]]
>>> list(findNeighbors(grid, 2, 1)) # find neighbors of 9
[8, 10, 5, 4, 6, 13, 12, 14]
>>> list(findNeighbors(grid, 3, 3)) # find neighbors of 15
[14, 11, 10]
For the sake of clarity, here is some equivalent code without all of the itertools magic:
def findNeighbors(grid, x, y):
if 0 < x < len(grid) - 1:
xi = (0, -1, 1) # this isn't first or last row, so we can look above and below
elif x > 0:
xi = (0, -1) # this is the last row, so we can only look above
else:
xi = (0, 1) # this is the first row, so we can only look below
# the following line accomplishes the same thing as the above code but for columns
yi = (0, -1, 1) if 0 < y < len(grid[0]) - 1 else ((0, -1) if y > 0 else (0, 1))
for a in xi:
for b in yi:
if a == b == 0: # this value is skipped using islice in the original code
continue
yield grid[x + a][y + b]