Mercurial > hg > numerics
annotate numerics/histogram.py @ 92:5b25e0be78aa
wip
| author | Jeff Hammel <k0scist@gmail.com> |
|---|---|
| date | Mon, 02 Mar 2015 16:08:39 -0800 |
| parents | bbe8f3e9615d |
| children | f6d885adb3d7 |
| rev | line source |
|---|---|
| 63 | 1 #!/usr/bin/env python |
| 2 # -*- coding: utf-8 -*- | |
| 3 | |
| 4 """ | |
| 68 | 5 Histograms |
| 63 | 6 |
| 7 Unicode is awesome; see http://www.alanwood.net/unicode/block_elements.html | |
| 8 """ | |
| 9 | |
| 10 blocks = """ | |
| 11 █ | |
| 12 ▉ | |
| 13 ▊ | |
| 14 ▋ | |
| 15 ▌ | |
| 16 ▍ | |
| 17 ▎ | |
| 18 ▏ | |
| 19 """ | |
| 20 | |
| 21 # imports | |
| 22 import argparse | |
| 23 import os | |
| 24 import sys | |
| 25 import time | |
| 75 | 26 from .data import transpose |
|
87
8448c62a4917
almost to output; frange done
Jeff Hammel <k0scist@gmail.com>
parents:
83
diff
changeset
|
27 from .generate import frange |
| 63 | 28 from .read import CSVParser |
| 92 | 29 from .write import CSVWriter |
| 64 | 30 from collections import OrderedDict |
| 63 | 31 |
| 32 # module globals | |
| 64 | 33 __all__ = ['Histogram', 'HistogramParser', 'main'] |
| 34 | |
| 35 class Histogram(object): | |
| 36 """historgram""" | |
| 63 | 37 |
| 64 | 38 def __init__(self, bins): |
| 73 | 39 self.bins = sorted(bins) |
| 64 | 40 assert len(bins) > 1 |
| 73 | 41 self.data = OrderedDict([(bin, []) |
| 42 for bin in zip(bins[:-1], | |
| 43 bins[1:])]) | |
| 64 | 44 |
| 70 | 45 def add(self, *values): |
| 46 """add values to the histogram""" | |
| 71 | 47 for value in values: |
| 72 | 48 for vmin, vmax in self.data.keys(): |
| 73 | 49 if vmin <= value < vmax: |
| 72 | 50 self.data[(vmin, vmax)].append(value) |
| 76 | 51 break |
| 52 else: | |
| 53 if value == vmax: | |
| 54 # handle rightmost endpoint | |
| 77 | 55 self.data[(vmin, vmax)].append(value) |
| 70 | 56 |
| 65 | 57 def __iadd__(self, value): |
| 71 | 58 self.add(value) |
| 65 | 59 return self |
| 60 | |
| 71 | 61 def __call__(self, *values): |
| 62 """ | |
| 63 add values to the histogram and return | |
| 64 OrderedDict of counts | |
| 65 """ | |
| 66 self.add(*values) | |
| 73 | 67 return OrderedDict([(bin, len(value)) |
| 68 for bin, value in self.data.items()]) | |
| 69 | |
| 70 def keys(self): | |
| 71 return self.data.keys() | |
| 71 | 72 |
| 74 | 73 def max(self): |
| 74 """return max length""" | |
| 75 return max([len(value) for value in self.data.values()]) | |
| 76 | |
| 77 | |
| 64 | 78 class HistogramParser(CSVParser): |
| 63 | 79 """histogram CLI option parser""" |
| 80 | |
| 81 def __init__(self, **kwargs): | |
| 82 kwargs.setdefault('formatter_class', argparse.RawTextHelpFormatter) | |
| 83 kwargs.setdefault('description', __doc__) | |
| 69 | 84 CSVParser.__init__(self, **kwargs) |
| 63 | 85 self.add_argument('-n', '--bins', dest='n_bins', type=int, |
| 86 help="number of bins") | |
| 76 | 87 self.add_argument('--min', dest='min', type=float, |
| 88 help="minimum value; else taken from data") | |
| 89 self.add_argument('--max', dest='max', type=float, | |
| 90 help="maximum value, else taken from data") | |
| 63 | 91 self.options = None |
| 92 | |
| 93 | |
| 94 def main(args=sys.argv[1:]): | |
| 95 """CLI""" | |
| 96 | |
| 97 # parse command line options | |
| 98 parser = HistogramParser() | |
| 99 options = parser.parse_args(args) | |
| 100 | |
| 75 | 101 # read data |
| 102 data = parser.read() | |
| 103 if not data: | |
| 104 parser.error("No data given") | |
| 105 | |
| 106 # transpose to columns | |
| 107 columns = transpose(data) | |
| 63 | 108 |
| 81 | 109 # cast to float |
| 82 | 110 columns = [[float(value) for value in column] |
| 111 for column in columns] | |
| 81 | 112 |
| 78 | 113 # find min, max if not provided |
| 114 if options.min is None: | |
| 115 options.min = min([min(column) for column in columns]) | |
| 116 if options.max is None: | |
| 89 | 117 options.max = max([max(column) for column in columns]) |
| 76 | 118 |
| 88 | 119 if not options.n_bins: |
| 83 | 120 # find number of bins, if not specified |
| 121 # We'll use a guess of 2 items per bin, on average | |
| 88 | 122 options.n_bins = len(columns[0]) / 2 |
| 83 | 123 |
|
87
8448c62a4917
almost to output; frange done
Jeff Hammel <k0scist@gmail.com>
parents:
83
diff
changeset
|
124 # make some bins |
| 88 | 125 bins = frange(options.min, options.max, options.n_bins) |
|
87
8448c62a4917
almost to output; frange done
Jeff Hammel <k0scist@gmail.com>
parents:
83
diff
changeset
|
126 |
| 83 | 127 # make some histograms |
|
87
8448c62a4917
almost to output; frange done
Jeff Hammel <k0scist@gmail.com>
parents:
83
diff
changeset
|
128 histograms = [] |
|
8448c62a4917
almost to output; frange done
Jeff Hammel <k0scist@gmail.com>
parents:
83
diff
changeset
|
129 for column in columns: |
|
8448c62a4917
almost to output; frange done
Jeff Hammel <k0scist@gmail.com>
parents:
83
diff
changeset
|
130 histogram = Histogram(bins) |
|
8448c62a4917
almost to output; frange done
Jeff Hammel <k0scist@gmail.com>
parents:
83
diff
changeset
|
131 histogram.add(*column) |
| 89 | 132 histograms.append(histogram) |
| 83 | 133 |
| 90 | 134 # record delimeters for output |
| 135 columns = zip(*histograms[0].keys()) | |
| 136 for histogram in histograms: | |
| 91 | 137 columns.append(histogram().values()) |
| 90 | 138 |
| 139 # output | |
| 140 writer = None # TODO | |
| 83 | 141 |
| 63 | 142 if __name__ == '__main__': |
| 143 main() |