[96fb8ad] | 1 | """Copyright (C) 2004 Paul Brossier <piem@altern.org> |
---|
| 2 | print aubio.__LICENSE__ for the terms of use |
---|
| 3 | """ |
---|
| 4 | |
---|
| 5 | __LICENSE__ = """\ |
---|
[20d8266] | 6 | Copyright (C) 2004-2009 Paul Brossier <piem@aubio.org> |
---|
[96fb8ad] | 7 | |
---|
[20d8266] | 8 | This file is part of aubio. |
---|
[96fb8ad] | 9 | |
---|
[20d8266] | 10 | aubio is free software: you can redistribute it and/or modify |
---|
| 11 | it under the terms of the GNU General Public License as published by |
---|
| 12 | the Free Software Foundation, either version 3 of the License, or |
---|
| 13 | (at your option) any later version. |
---|
| 14 | |
---|
| 15 | aubio is distributed in the hope that it will be useful, |
---|
| 16 | but WITHOUT ANY WARRANTY; without even the implied warranty of |
---|
| 17 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
---|
| 18 | GNU General Public License for more details. |
---|
| 19 | |
---|
| 20 | You should have received a copy of the GNU General Public License |
---|
| 21 | along with aubio. If not, see <http://www.gnu.org/licenses/>. |
---|
[96fb8ad] | 22 | """ |
---|
| 23 | |
---|
| 24 | """ |
---|
| 25 | original author Tim Peters |
---|
| 26 | modified by Paul Brossier <piem@altern.org> |
---|
| 27 | inspired from http://www.ics.uci.edu/~eppstein/161/python/peters-selection.py |
---|
| 28 | """ |
---|
| 29 | |
---|
| 30 | def short_find(a, rank): |
---|
[0484dc1] | 31 | """ find the rank-th value in sorted a """ |
---|
| 32 | # copy to b before sorting |
---|
| 33 | b = a[:] |
---|
| 34 | b.sort() |
---|
| 35 | return b[rank - 1] |
---|
[96fb8ad] | 36 | |
---|
| 37 | def percental(a, rank): |
---|
[0484dc1] | 38 | """ Find the rank'th-smallest value in a, in worst-case linear time. """ |
---|
[96fb8ad] | 39 | n = len(a) |
---|
| 40 | assert 1 <= rank <= n |
---|
| 41 | if n <= 7: |
---|
| 42 | return short_find(a, rank) |
---|
| 43 | |
---|
| 44 | ## Find median of median-of-7's. |
---|
| 45 | ##medians = [short_find(a[i : i+7], 4) for i in xrange(0, n-6, 7)] |
---|
| 46 | #median = find(medians, (len(medians) + 1) // 2) |
---|
| 47 | |
---|
| 48 | # modified to Find median |
---|
| 49 | median = short_find([a[0], a[-1], a[n//2]], 2) |
---|
| 50 | |
---|
| 51 | # Partition around the median. |
---|
| 52 | # a[:i] <= median |
---|
| 53 | # a[j+1:] >= median |
---|
| 54 | i, j = 0, n-1 |
---|
| 55 | while i <= j: |
---|
| 56 | while a[i] < median: |
---|
| 57 | i += 1 |
---|
| 58 | while a[j] > median: |
---|
| 59 | j -= 1 |
---|
| 60 | if i <= j: |
---|
| 61 | a[i], a[j] = a[j], a[i] |
---|
| 62 | i += 1 |
---|
| 63 | j -= 1 |
---|
| 64 | |
---|
| 65 | if rank <= i: |
---|
| 66 | return percental(a[:i], rank) |
---|
| 67 | else: |
---|
| 68 | return percental(a[i:], rank - i) |
---|
| 69 | |
---|