summaryrefslogtreecommitdiff
path: root/util/stats/info.py
blob: ae5d3211f1bc4a1ba3e70a2be02e6dea9775b4a8 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
# Copyright (c) 2003-2004 The Regents of The University of Michigan
# All rights reserved.
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions are
# met: redistributions of source code must retain the above copyright
# notice, this list of conditions and the following disclaimer;
# redistributions in binary form must reproduce the above copyright
# notice, this list of conditions and the following disclaimer in the
# documentation and/or other materials provided with the distribution;
# neither the name of the copyright holders nor the names of its
# contributors may be used to endorse or promote products derived from
# this software without specific prior written permission.
#
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

from __future__ import division
import operator, re, types

source = None
display_run = 0
global globalTicks
globalTicks = None

def total(f):
    if isinstance(f, FormulaStat):
        v = f.value
    else:
        v = f

    f = FormulaStat()
    if isinstance(v, (list, tuple)):
        f.value = reduce(operator.add, v)
    else:
        f.value = v

    return f

def unaryop(op, f):
    if isinstance(f, FormulaStat):
        v = f.value
    else:
        v = f

    if isinstance(v, (list, tuple)):
        return map(op, v)
    else:
        return op(v)

def zerodiv(lv, rv):
    if rv == 0.0:
        return 0.0
    else:
        return operator.truediv(lv, rv)

def wrapop(op, lv, rv):
    if isinstance(lv, str):
        return lv

    if isinstance(rv, str):
        return rv

    return op(lv, rv)

def same(lrun, rrun):
    for lx,rx in zip(lrun.keys(),rrun.keys()):
        if lx != rx:
            print 'lx != rx'
            print lx, rx
            print lrun.keys()
            print rrun.keys()
            return False
        for ly,ry in zip(lrun[lx].keys(),rrun[rx].keys()):
            if ly != ry:
                print 'ly != ry'
                print ly, ry
                print lrun[lx].keys()
                print rrun[rx].keys()
                return False
    return True


def binaryop(op, lf, rf):
    result = {}

    if isinstance(lf, FormulaStat) and isinstance(rf, FormulaStat):
        lv = lf.value
        rv = rf.value

        theruns = []
        for r in lv.keys():
            if rv.has_key(r):
                if same(lv[r], rv[r]):
                    theruns.append(r)
                else:
                    raise AttributeError

        for run in theruns:
            result[run] = {}
            for x in lv[run].keys():
                result[run][x] = {}
                for y in lv[run][x].keys():
                    result[run][x][y] = wrapop(op, lv[run][x][y],
                                               rv[run][x][y])
    elif isinstance(lf, FormulaStat):
        lv = lf.value
        for run in lv.keys():
            result[run] = {}
            for x in lv[run].keys():
                result[run][x] = {}
                for y in lv[run][x].keys():
                    result[run][x][y] = wrapop(op, lv[run][x][y], rf)
    elif isinstance(rf, FormulaStat):
        rv = rf.value
        for run in rv.keys():
            result[run] = {}
            for x in rv[run].keys():
                result[run][x] = {}
                for y in rv[run][x].keys():
                    result[run][x][y] = wrapop(op, lf, rv[run][x][y])

    return result

def sums(x, y):
    if isinstance(x, (list, tuple)):
        return map(lambda x, y: x + y, x, y)
    else:
        return x + y

def alltrue(seq):
    return reduce(lambda x, y: x and y, seq)

def allfalse(seq):
    return not reduce(lambda x, y: x or y, seq)

def enumerate(seq):
    return map(None, range(len(seq)), seq)

def cmp(a, b):
    if a < b:
        return -1
    elif a == b:
        return 0
    else:
        return 1

class Statistic(object):

    def __init__(self, data):
        self.__dict__.update(data.__dict__)
        if not self.__dict__.has_key('value'):
            self.__dict__['value'] = None
        if not self.__dict__.has_key('bins'):
            self.__dict__['bins'] = None
        if not self.__dict__.has_key('ticks'):
            self.__dict__['ticks'] = None
        if 'vc' not in self.__dict__:
            self.vc = {}

    def __getattribute__(self, attr):
        if attr == 'ticks':
            if self.__dict__['ticks'] != globalTicks:
                self.__dict__['value'] = None
                self.__dict__['ticks'] = globalTicks
            return self.__dict__['ticks']
        if attr == 'value':
            if self.__dict__['ticks'] != globalTicks:
                if self.__dict__['ticks'] != None and \
                                    len(self.__dict__['ticks']) == 1:
                    self.vc[self.__dict__['ticks'][0]] = self.__dict__['value']
                self.__dict__['ticks'] = globalTicks
                if len(globalTicks) == 1 and self.vc.has_key(globalTicks[0]):
                    self.__dict__['value'] = self.vc[globalTicks[0]]
                else:
                    self.__dict__['value'] = None
            if self.__dict__['value'] == None:
                self.__dict__['value'] = self.getValue()
            return self.__dict__['value']
        else:
            return super(Statistic, self).__getattribute__(attr)

    def __setattr__(self, attr, value):
        if attr == 'bins' or attr == 'ticks':
            if attr == 'bins':
                if value is not None:
                    value = source.getBin(value)
            #elif attr == 'ticks' and type(value) is str:
            #    value = [ int(x) for x in value.split() ]

            self.__dict__[attr] = value
            self.__dict__['value'] = None
            self.vc = {}
        else:
            super(Statistic, self).__setattr__(attr, value)

    def getValue(self):
        raise AttributeError, 'getValue() must be defined'

    def zero(self):
        return False

    def __ne__(self, other):
        return not (self == other)

    def __str__(self):
        return '%f' % (float(self))

class FormulaStat(object):
    def __add__(self, other):
        f = FormulaStat()
        f.value = binaryop(operator.add, self, other)
        return f
    def __sub__(self, other):
        f = FormulaStat()
        f.value = binaryop(operator.sub, self, other)
        return f
    def __mul__(self, other):
        f = FormulaStat()
        f.value = binaryop(operator.mul, self, other)
        return f
    def __truediv__(self, other):
        f = FormulaStat()
        f.value = binaryop(zerodiv, self, other)
        return f
    def __mod__(self, other):
        f = FormulaStat()
        f.value = binaryop(operator.mod, self, other)
        return f
    def __radd__(self, other):
        f = FormulaStat()
        f.value = binaryop(operator.add, other, self)
        return f
    def __rsub__(self, other):
        f = FormulaStat()
        f.value = binaryop(operator.sub, other, self)
        return f
    def __rmul__(self, other):
        f = FormulaStat()
        f.value = binaryop(operator.mul, other, self)
        return f
    def __rtruediv__(self, other):
        f = FormulaStat()
        f.value = binaryop(zerodiv, other, self)
        return f
    def __rmod__(self, other):
        f = FormulaStat()
        f.value = binaryop(operator.mod, other, self)
        return f
    def __neg__(self):
        f = FormulaStat()
        f.value = unaryop(operator.neg, self)
        return f
    def __getitem__(self, idx):
        f = FormulaStat()
        f.value = {}
        for key in self.value.keys():
            f.value[key] = {}
            f.value[key][0] = {}
            f.value[key][0][0] = self.value[key][idx][0]
        return f

    def __float__(self):
        if isinstance(self.value, FormulaStat):
            return float(self.value)
        if not self.value.has_key(display_run):
            return (1e300*1e300)
        if len(self.value[display_run]) == 1:
            return self.value[display_run][0][0]
        else:
            #print self.value[display_run]
            return self.value[display_run][4][0]
            #raise ValueError

    def display(self):
        import display
        d = display.VectorDisplay()
        d.flags = 0
        d.precision = 1
        d.name = 'formula'
        d.desc = 'formula'
        val = self.value[display_run]
        d.value = [ val[x][0] for x in val.keys() ]
        d.display()


class Scalar(Statistic,FormulaStat):
    def getValue(self):
        return source.data(self, self.bins, self.ticks)

    def display(self):
        import display
        p = display.Print()
        p.name = self.name
        p.desc = self.desc
        p.value = float(self)
        p.flags = self.flags
        p.precision = self.precision
        if display.all or (self.flags & flags.printable):
            p.display()

    def comparable(self, other):
        return self.name == other.name

    def __eq__(self, other):
        return self.value == other.value

    def __isub__(self, other):
        self.value -= other.value
        return self

    def __iadd__(self, other):
        self.value += other.value
        return self

    def __itruediv__(self, other):
        if not other:
            return self
        self.value /= other
        return self

class Vector(Statistic,FormulaStat):
    def getValue(self):
        return source.data(self, self.bins, self.ticks);

    def display(self):
        import display
        if not display.all and not (self.flags & flags.printable):
            return

        d = display.VectorDisplay()
        d.__dict__.update(self.__dict__)
        d.display()

    def comparable(self, other):
        return self.name == other.name and \
               len(self.value) == len(other.value)

    def __eq__(self, other):
        if isinstance(self.value, (list, tuple)) != \
               isinstance(other.value, (list, tuple)):
            return False

        if isinstance(self.value, (list, tuple)):
            if len(self.value) != len(other.value):
                return False
            else:
                for v1,v2 in zip(self.value, other.value):
                    if v1 != v2:
                        return False
                return True
        else:
            return self.value == other.value

    def __isub__(self, other):
        self.value = binaryop(operator.sub, self.value, other.value)
        return self

    def __iadd__(self, other):
        self.value = binaryop(operator.add, self.value, other.value)
        return self

    def __itruediv__(self, other):
        if not other:
            return self
        if isinstance(self.value, (list, tuple)):
            for i in xrange(len(self.value)):
                self.value[i] /= other
        else:
            self.value /= other
        return self

class Formula(Vector):
    def getValue(self):
        formula = re.sub(':', '__', self.formula)
        x = eval(formula, source.stattop)
        return x.value

    def comparable(self, other):
        return self.name == other.name and \
               compare(self.dist, other.dist)

    def __eq__(self, other):
        return self.value == other.value

    def __isub__(self, other):
        return self

    def __iadd__(self, other):
        return self

    def __itruediv__(self, other):
        if not other:
            return self
        return self

class SimpleDist(object):
    def __init__(self, sums, squares, samples):
        self.sums = sums
        self.squares = squares
        self.samples = samples

    def getValue(self):
        return 0.0

    def display(self, name, desc, flags, precision):
        import display
        p = display.Print()
        p.flags = flags
        p.precision = precision

        if self.samples > 0:
            p.name = name + ".mean"
            p.value = self.sums / self.samples
            p.display()

            p.name = name + ".stdev"
            if self.samples > 1:
                var = (self.samples * self.squares - self.sums ** 2) \
                      / (self.samples * (self.samples - 1))
                if var >= 0:
                    p.value = math.sqrt(var)
                else:
                    p.value = 'NaN'
            else:
                p.value = 0.0
            p.display()

        p.name = name + ".samples"
        p.value = self.samples
        p.display()

    def comparable(self, other):
        return True

    def __eq__(self, other):
        return self.sums == other.sums and self.squares == other.squares and \
               self.samples == other.samples

    def __isub__(self, other):
        self.sums -= other.sums
        self.squares -= other.squares
        self.samples -= other.samples
        return self

    def __iadd__(self, other):
        self.sums += other.sums
        self.squares += other.squares
        self.samples += other.samples
        return self

    def __itruediv__(self, other):
        if not other:
            return self
        self.sums /= other
        self.squares /= other
        self.samples /= other
        return self

class FullDist(SimpleDist):
    def __init__(self, sums, squares, samples, minval, maxval,
                 under, vec, over, min, max, bsize, size):
        self.sums = sums
        self.squares = squares
        self.samples = samples
        self.minval = minval
        self.maxval = maxval
        self.under = under
        self.vec = vec
        self.over = over
        self.min = min
        self.max = max
        self.bsize = bsize
        self.size = size

    def getValue(self):
        return 0.0

    def display(self, name, desc, flags, precision):
        import display
        p = display.Print()
        p.flags = flags
        p.precision = precision

        p.name = name + '.min_val'
        p.value = self.minval
        p.display()

        p.name = name + '.max_val'
        p.value = self.maxval
        p.display()

        p.name = name + '.underflow'
        p.value = self.under
        p.display()

        i = self.min
        for val in self.vec[:-1]:
            p.name = name + '[%d:%d]' % (i, i + self.bsize - 1)
            p.value = val
            p.display()
            i += self.bsize

        p.name = name + '[%d:%d]' % (i, self.max)
        p.value = self.vec[-1]
        p.display()


        p.name = name + '.overflow'
        p.value = self.over
        p.display()

        SimpleDist.display(self, name, desc, flags, precision)

    def comparable(self, other):
        return self.min == other.min and self.max == other.max and \
               self.bsize == other.bsize and self.size == other.size

    def __eq__(self, other):
        return self.sums == other.sums and self.squares == other.squares and \
               self.samples == other.samples

    def __isub__(self, other):
        self.sums -= other.sums
        self.squares -= other.squares
        self.samples -= other.samples

        if other.samples:
            self.minval = min(self.minval, other.minval)
            self.maxval = max(self.maxval, other.maxval)
            self.under -= under
            self.vec = map(lambda x,y: x - y, self.vec, other.vec)
            self.over -= over
        return self

    def __iadd__(self, other):
        if not self.samples and other.samples:
            self = other
            return self

        self.sums += other.sums
        self.squares += other.squares
        self.samples += other.samples

        if other.samples:
            self.minval = min(self.minval, other.minval)
            self.maxval = max(self.maxval, other.maxval)
            self.under += other.under
            self.vec = map(lambda x,y: x + y, self.vec, other.vec)
            self.over += other.over
        return self

    def __itruediv__(self, other):
        if not other:
            return self
        self.sums /= other
        self.squares /= other
        self.samples /= other

        if self.samples:
            self.under /= other
            for i in xrange(len(self.vec)):
                self.vec[i] /= other
            self.over /= other
        return self

class Dist(Statistic):
    def getValue(self):
        return 0.0

    def display(self):
        import display
        if not display.all and not (self.flags & flags.printable):
            return

        self.dist.display(self.name, self.desc, self.flags, self.precision)

    def comparable(self, other):
        return self.name == other.name and \
               self.dist.compareable(other.dist)

    def __eq__(self, other):
        return self.dist == other.dist

    def __isub__(self, other):
        self.dist -= other.dist
        return self

    def __iadd__(self, other):
        self.dist += other.dist
        return self

    def __itruediv__(self, other):
        if not other:
            return self
        self.dist /= other
        return self

class VectorDist(Statistic):
    def getValue(self):
        return 0.0

    def display(self):
        import display
        if not display.all and not (self.flags & flags.printable):
            return

        if isinstance(self.dist, SimpleDist):
            return

        for dist,sn,sd,i in map(None, self.dist, self.subnames, self.subdescs,
                                range(len(self.dist))):
            if len(sn) > 0:
                name = '%s.%s' % (self.name, sn)
            else:
                name = '%s[%d]' % (self.name, i)

            if len(sd) > 0:
                desc = sd
            else:
                desc = self.desc

            dist.display(name, desc, self.flags, self.precision)

        if (self.flags & flags.total) or 1:
            if isinstance(self.dist[0], SimpleDist):
                disttotal = SimpleDist( \
                    reduce(sums, [d.sums for d in self.dist]),
                    reduce(sums, [d.squares for d in self.dist]),
                    reduce(sums, [d.samples for d in self.dist]))
            else:
                disttotal = FullDist( \
                    reduce(sums, [d.sums for d in self.dist]),
                    reduce(sums, [d.squares for d in self.dist]),
                    reduce(sums, [d.samples for d in self.dist]),
                    min([d.minval for d in self.dist]),
                    max([d.maxval for d in self.dist]),
                    reduce(sums, [d.under for d in self.dist]),
                    reduce(sums, [d.vec for d in self.dist]),
                    reduce(sums, [d.over for d in self.dist]),
                    dist[0].min,
                    dist[0].max,
                    dist[0].bsize,
                    dist[0].size)

            name = '%s.total' % (self.name)
            desc = self.desc
            disttotal.display(name, desc, self.flags, self.precision)

    def comparable(self, other):
        return self.name == other.name and \
               alltrue(map(lambda x, y : x.comparable(y),
                           self.dist,
                           other.dist))

    def __eq__(self, other):
        return alltrue(map(lambda x, y : x == y, self.dist, other.dist))

    def __isub__(self, other):
        if isinstance(self.dist, (list, tuple)) and \
               isinstance(other.dist, (list, tuple)):
            for sd,od in zip(self.dist, other.dist):
                sd -= od
        else:
            self.dist -= other.dist
        return self

    def __iadd__(self, other):
        if isinstance(self.dist, (list, tuple)) and \
               isinstance(other.dist, (list, tuple)):
            for sd,od in zip(self.dist, other.dist):
                sd += od
        else:
            self.dist += other.dist
        return self

    def __itruediv__(self, other):
        if not other:
            return self
        if isinstance(self.dist, (list, tuple)):
            for dist in self.dist:
                dist /= other
        else:
            self.dist /= other
        return self

class Vector2d(Statistic):
    def getValue(self):
        return 0.0

    def display(self):
        import display
        if not display.all and not (self.flags & flags.printable):
            return

        d = display.VectorDisplay()
        d.__dict__.update(self.__dict__)

        if self.__dict__.has_key('ysubnames'):
            ysubnames = list(self.ysubnames)
            slack = self.x - len(ysubnames)
            if slack > 0:
                ysubnames.extend(['']*slack)
        else:
            ysubnames = range(self.x)

        for x,sname in enumerate(ysubnames):
            o = x * self.y
            d.value = self.value[o:o+self.y]
            d.name = '%s[%s]' % (self.name, sname)
            d.display()

        if self.flags & flags.total:
            d.value = []
            for y in range(self.y):
                xtot = 0.0
                for x in range(self.x):
                    xtot += self.value[y + x * self.x]
                d.value.append(xtot)

            d.name = self.name + '.total'
            d.display()

    def comparable(self, other):
        return self.name == other.name and self.x == other.x and \
               self.y == other.y

    def __eq__(self, other):
        return True

    def __isub__(self, other):
        return self

    def __iadd__(self, other):
        return self

    def __itruediv__(self, other):
        if not other:
            return self
        return self

def NewStat(data):
    stat = None
    if data.type == 'SCALAR':
        stat = Scalar(data)
    elif data.type == 'VECTOR':
        stat = Vector(data)
    elif data.type == 'DIST':
        stat = Dist(data)
    elif data.type == 'VECTORDIST':
        stat = VectorDist(data)
    elif data.type == 'VECTOR2D':
        stat = Vector2d(data)
    elif data.type == 'FORMULA':
        stat = Formula(data)

    return stat