Reputation: 485
I have joined Standford online course on Algorithms design and now I'm solving the first programming question.
The file contains all the 100,000 integers between 1 and 100,000 (including both) in some random order(no integer is repeated). Your task is to find the number of inversions in the file given (every row has a single integer between 1 and 100,000). Assume your array is from 1 to 100,000 and i-th row of the file gives you the i-th entry of the array.
Update: I have found that my code works only for the 2^n case. Problem is in the code, not Python. I have updated the code, now it work fine and I have done the quiz. Thanks to all who helped
Fixed code is:
def merge_count_split (a, b):
c = []
inv = 0
i=0
j=0
for k in range( len(a) + len(b) ):
if i < len(a) and j < len(b):
if a[i] < b[j]:
c.append(a[i])
i += 1
elif a[i] > b[j]:
c.append(b[j])
inv += len(a)-i
j += 1
elif i == len(a):
c.append(b[j])
j += 1
elif j == len(b):
c.append(a[i])
i += 1
return c, inv
def count_inv (data):
n = len(data)
if n == 1:
return data, 0
a, x = count_inv(data[:n/2])
b, y = count_inv(data[n/2:])
c, z = merge_count_split(a,b)
return c, x + y + z
with open('IntegerArray.txt') as f:
array = [int(line) for line in f]
print count_inv(array)[0]
This program works fine for small arrays, but for the large array from the question it prints array of 65536 numbers in proper order, not 100000, as I expect. It omits numbers at random places.
What is the reason for this unexpected behaviour of python?
Upvotes: 2
Views: 2960
Reputation: 150987
By setting n = len(a)
and merging only n * 2
times, you truncate b
if it's longer than a
.
This partially explains the striking fact that you have 2 ** 16 items in the resulting list.
Upvotes: 2