1Primitive(3)          User Contributed Perl Documentation         Primitive(3)
2
3
4

NAME

6       PDL::Primitive - primitive operations for pdl
7

DESCRIPTION

9       This module provides some primitive and useful functions defined using
10       PDL::PP and able to use the new indexing tricks.
11
12       See PDL::Indexing for how to use indices creatively.  For explanation
13       of the signature format, see PDL::PP.
14

SYNOPSIS

16        # Pulls in PDL::Primitive, among other modules.
17        use PDL;
18
19        # Only pull in PDL::Primitive:
20        use PDL::Primitive;
21

FUNCTIONS

23   inner
24         Signature: (a(n); b(n); [o]c())
25
26       Inner product over one dimension
27
28        c = sum_i a_i * b_i
29
30       If "a() * b()" contains only bad data, "c()" is set bad. Otherwise
31       "c()" will have its bad flag cleared, as it will not contain any bad
32       values.
33
34   outer
35         Signature: (a(n); b(m); [o]c(n,m))
36
37       outer product over one dimension
38
39       Naturally, it is possible to achieve the effects of outer product
40       simply by threading over the ""*"" operator but this function is
41       provided for convenience.
42
43       outer processes bad values.  It will set the bad-value flag of all
44       output piddles if the flag is set for any of the input piddles.
45
46   x
47        Signature: (a(i,z), b(x,i),[o]c(x,z))
48
49       Matrix multiplication
50
51       PDL overloads the "x" operator (normally the repeat operator) for
52       matrix multiplication.  The number of columns (size of the 0 dimension)
53       in the left-hand argument must normally equal the number of rows (size
54       of the 1 dimension) in the right-hand argument.
55
56       Row vectors are represented as (N x 1) two-dimensional PDLs, or you may
57       be sloppy and use a one-dimensional PDL.  Column vectors are
58       represented as (1 x N) two-dimensional PDLs.
59
60       Threading occurs in the usual way, but as both the 0 and 1 dimension
61       (if present) are included in the operation, you must be sure that you
62       don't try to thread over either of those dims.
63
64       EXAMPLES
65
66       Here are some simple ways to define vectors and matrices:
67
68        pdl> $r = pdl(1,2);                # A row vector
69        pdl> $c = pdl([[3],[4]]);          # A column vector
70        pdl> $c = pdl(3,4)->(*1);          # A column vector, using NiceSlice
71        pdl> $m = pdl([[1,2],[3,4]]);      # A 2x2 matrix
72
73       Now that we have a few objects prepared, here is how to matrix-multiply
74       them:
75
76        pdl> print $r x $m                 # row x matrix = row
77        [
78         [ 7 10]
79        ]
80
81        pdl> print $m x $r                 # matrix x row = ERROR
82        PDL: Dim mismatch in matmult of [2x2] x [2x1]: 2 != 1
83
84        pdl> print $m x $c                 # matrix x column = column
85        [
86         [ 5]
87         [11]
88        ]
89
90        pdl> print $m x 2                  # Trivial case: scalar mult.
91        [
92         [2 4]
93         [6 8]
94        ]
95
96        pdl> print $r x $c                 # row x column = scalar
97        [
98         [11]
99        ]
100
101        pdl> print $c x $r                 # column x row = matrix
102        [
103         [3 6]
104         [4 8]
105        ]
106
107       INTERNALS
108
109       The mechanics of the multiplication are carried out by the matmult
110       method.
111
112   matmult
113         Signature: (a(t,h); b(w,t); [o]c(w,h))
114
115       Matrix multiplication
116
117       Notionally, matrix multiplication $x x $y is equivalent to the
118       threading expression
119
120           $x->dummy(1)->inner($y->xchg(0,1)->dummy(2),$c);
121
122       but for large matrices that breaks CPU cache and is slow.  Instead,
123       matmult calculates its result in 32x32x32 tiles, to keep the memory
124       footprint within cache as long as possible on most modern CPUs.
125
126       For usage, see x, a description of the overloaded 'x' operator
127
128       matmult ignores the bad-value flag of the input piddles.  It will set
129       the bad-value flag of all output piddles if the flag is set for any of
130       the input piddles.
131
132   innerwt
133         Signature: (a(n); b(n); c(n); [o]d())
134
135       Weighted (i.e. triple) inner product
136
137        d = sum_i a(i) b(i) c(i)
138
139       innerwt processes bad values.  It will set the bad-value flag of all
140       output piddles if the flag is set for any of the input piddles.
141
142   inner2
143         Signature: (a(n); b(n,m); c(m); [o]d())
144
145       Inner product of two vectors and a matrix
146
147        d = sum_ij a(i) b(i,j) c(j)
148
149       Note that you should probably not thread over "a" and "c" since that
150       would be very wasteful. Instead, you should use a temporary for "b*c".
151
152       inner2 processes bad values.  It will set the bad-value flag of all
153       output piddles if the flag is set for any of the input piddles.
154
155   inner2d
156         Signature: (a(n,m); b(n,m); [o]c())
157
158       Inner product over 2 dimensions.
159
160       Equivalent to
161
162        $c = inner($x->clump(2), $y->clump(2))
163
164       inner2d processes bad values.  It will set the bad-value flag of all
165       output piddles if the flag is set for any of the input piddles.
166
167   inner2t
168         Signature: (a(j,n); b(n,m); c(m,k); [t]tmp(n,k); [o]d(j,k)))
169
170       Efficient Triple matrix product "a*b*c"
171
172       Efficiency comes from by using the temporary "tmp". This operation only
173       scales as "N**3" whereas threading using inner2 would scale as "N**4".
174
175       The reason for having this routine is that you do not need to have the
176       same thread-dimensions for "tmp" as for the other arguments, which in
177       case of large numbers of matrices makes this much more memory-
178       efficient.
179
180       It is hoped that things like this could be taken care of as a kind of
181       closures at some point.
182
183       inner2t processes bad values.  It will set the bad-value flag of all
184       output piddles if the flag is set for any of the input piddles.
185
186   crossp
187         Signature: (a(tri=3); b(tri); [o] c(tri))
188
189       Cross product of two 3D vectors
190
191       After
192
193        $c = crossp $x, $y
194
195       the inner product "$c*$x" and "$c*$y" will be zero, i.e. $c is
196       orthogonal to $x and $y
197
198       crossp does not process bad values.  It will set the bad-value flag of
199       all output piddles if the flag is set for any of the input piddles.
200
201   norm
202         Signature: (vec(n); [o] norm(n))
203
204       Normalises a vector to unit Euclidean length
205
206       norm processes bad values.  It will set the bad-value flag of all
207       output piddles if the flag is set for any of the input piddles.
208
209   indadd
210         Signature: (a(); indx ind(); [o] sum(m))
211
212       Threaded Index Add: Add "a" to the "ind" element of "sum", i.e:
213
214        sum(ind) += a
215
216       Simple Example:
217
218         $x = 2;
219         $ind = 3;
220         $sum = zeroes(10);
221         indadd($x,$ind, $sum);
222         print $sum
223         #Result: ( 2 added to element 3 of $sum)
224         # [0 0 0 2 0 0 0 0 0 0]
225
226       Threaded Example:
227
228         $x = pdl( 1,2,3);
229         $ind = pdl( 1,4,6);
230         $sum = zeroes(10);
231         indadd($x,$ind, $sum);
232         print $sum."\n";
233         #Result: ( 1, 2, and 3 added to elements 1,4,6 $sum)
234         # [0 1 0 0 2 0 3 0 0 0]
235
236       The routine barfs if any of the indices are bad.
237
238   conv1d
239         Signature: (a(m); kern(p); [o]b(m); int reflect)
240
241       1D convolution along first dimension
242
243       The m-th element of the discrete convolution of an input piddle $a of
244       size $M, and a kernel piddle $kern of size $P, is calculated as
245
246                                     n = ($P-1)/2
247                                     ====
248                                     \
249         ($a conv1d $kern)[m]   =     >      $a_ext[m - n] * $kern[n]
250                                     /
251                                     ====
252                                     n = -($P-1)/2
253
254       where $a_ext is either the periodic (or reflected) extension of $a so
255       it is equal to $a on " 0..$M-1 " and equal to the corresponding
256       periodic/reflected image of $a outside that range.
257
258         $con = conv1d sequence(10), pdl(-1,0,1);
259
260         $con = conv1d sequence(10), pdl(-1,0,1), {Boundary => 'reflect'};
261
262       By default, periodic boundary conditions are assumed (i.e. wrap
263       around).  Alternatively, you can request reflective boundary conditions
264       using the "Boundary" option:
265
266         {Boundary => 'reflect'} # case in 'reflect' doesn't matter
267
268       The convolution is performed along the first dimension. To apply it
269       across another dimension use the slicing routines, e.g.
270
271         $y = $x->mv(2,0)->conv1d($kernel)->mv(0,2); # along third dim
272
273       This function is useful for threaded filtering of 1D signals.
274
275       Compare also conv2d, convolve, fftconvolve, fftwconv, rfftwconv
276
277       WARNING: "conv1d" processes bad values in its inputs as the numeric
278       value of "$pdl->badvalue" so it is not recommended for processing pdls
279       with bad values in them unless special care is taken.
280
281       conv1d ignores the bad-value flag of the input piddles.  It will set
282       the bad-value flag of all output piddles if the flag is set for any of
283       the input piddles.
284
285   in
286         Signature: (a(); b(n); [o] c())
287
288       test if a is in the set of values b
289
290          $goodmsk = $labels->in($goodlabels);
291          print pdl(3,1,4,6,2)->in(pdl(2,3,3));
292         [1 0 0 0 1]
293
294       "in" is akin to the is an element of of set theory. In principle, PDL
295       threading could be used to achieve its functionality by using a
296       construct like
297
298          $msk = ($labels->dummy(0) == $goodlabels)->orover;
299
300       However, "in" doesn't create a (potentially large) intermediate and is
301       generally faster.
302
303       in does not process bad values.  It will set the bad-value flag of all
304       output piddles if the flag is set for any of the input piddles.
305
306   uniq
307       return all unique elements of a piddle
308
309       The unique elements are returned in ascending order.
310
311         PDL> p pdl(2,2,2,4,0,-1,6,6)->uniq
312         [-1 0 2 4 6]     # 0 is returned 2nd (sorted order)
313
314         PDL> p pdl(2,2,2,4,nan,-1,6,6)->uniq
315         [-1 2 4 6 nan]   # NaN value is returned at end
316
317       Note: The returned pdl is 1D; any structure of the input piddle is
318       lost.  "NaN" values are never compare equal to any other values, even
319       themselves.  As a result, they are always unique. "uniq" returns the
320       NaN values at the end of the result piddle.  This follows the Matlab
321       usage.
322
323       See uniqind if you need the indices of the unique elements rather than
324       the values.
325
326       Bad values are not considered unique by uniq and are ignored.
327
328        $x=sequence(10);
329        $x=$x->setbadif($x%3);
330        print $x->uniq;
331        [0 3 6 9]
332
333   uniqind
334       Return the indices of all unique elements of a piddle The order is in
335       the order of the values to be consistent with uniq. "NaN" values never
336       compare equal with any other value and so are always unique.  This
337       follows the Matlab usage.
338
339         PDL> p pdl(2,2,2,4,0,-1,6,6)->uniqind
340         [5 4 1 3 6]     # the 0 at index 4 is returned 2nd, but...
341
342         PDL> p pdl(2,2,2,4,nan,-1,6,6)->uniqind
343         [5 1 3 6 4]     # ...the NaN at index 4 is returned at end
344
345       Note: The returned pdl is 1D; any structure of the input piddle is
346       lost.
347
348       See uniq if you want the unique values instead of the indices.
349
350       Bad values are not considered unique by uniqind and are ignored.
351
352   uniqvec
353       Return all unique vectors out of a collection
354
355         NOTE: If any vectors in the input piddle have NaN values
356         they are returned at the end of the non-NaN ones.  This is
357         because, by definition, NaN values never compare equal with
358         any other value.
359
360         NOTE: The current implementation does not sort the vectors
361         containing NaN values.
362
363       The unique vectors are returned in lexicographically sorted ascending
364       order. The 0th dimension of the input PDL is treated as a dimensional
365       index within each vector, and the 1st and any higher dimensions are
366       taken to run across vectors. The return value is always 2D; any
367       structure of the input PDL (beyond using the 0th dimension for vector
368       index) is lost.
369
370       See also uniq for a unique list of scalars; and qsortvec for sorting a
371       list of vectors lexicographcally.
372
373       If a vector contains all bad values, it is ignored as in uniq.  If some
374       of the values are good, it is treated as a normal vector. For example,
375       [1 2 BAD] and [BAD 2 3] could be returned, but [BAD BAD BAD] could not.
376       Vectors containing BAD values will be returned after any non-NaN and
377       non-BAD containing vectors, followed by the NaN vectors.
378
379   hclip
380         Signature: (a(); b(); [o] c())
381
382       clip (threshold) $a by $b ($b is upper bound)
383
384       hclip processes bad values.  It will set the bad-value flag of all
385       output piddles if the flag is set for any of the input piddles.
386
387   lclip
388         Signature: (a(); b(); [o] c())
389
390       clip (threshold) $a by $b ($b is lower bound)
391
392       lclip processes bad values.  It will set the bad-value flag of all
393       output piddles if the flag is set for any of the input piddles.
394
395   clip
396       Clip (threshold) a piddle by (optional) upper or lower bounds.
397
398        $y = $x->clip(0,3);
399        $c = $x->clip(undef, $x);
400
401       clip handles bad values since it is just a wrapper around hclip and
402       lclip.
403
404   clip
405         Signature: (a(); l(); h(); [o] c())
406
407       info not available
408
409       clip processes bad values.  It will set the bad-value flag of all
410       output piddles if the flag is set for any of the input piddles.
411
412   wtstat
413         Signature: (a(n); wt(n); avg(); [o]b(); int deg)
414
415       Weighted statistical moment of given degree
416
417       This calculates a weighted statistic over the vector "a".  The formula
418       is
419
420        b() = (sum_i wt_i * (a_i ** degree - avg)) / (sum_i wt_i)
421
422       Bad values are ignored in any calculation; $b will only have its bad
423       flag set if the output contains any bad data.
424
425   statsover
426         Signature: (a(n); w(n); float+ [o]avg(); float+ [o]prms(); int+ [o]median(); int+ [o]min(); int+ [o]max(); float+ [o]adev(); float+ [o]rms())
427
428       Calculate useful statistics over a dimension of a piddle
429
430         ($mean,$prms,$median,$min,$max,$adev,$rms) = statsover($piddle, $weights);
431
432       This utility function calculates various useful quantities of a piddle.
433       These are:
434
435       ·  the mean:
436
437            MEAN = sum (x)/ N
438
439          with "N" being the number of elements in x
440
441       ·  the population RMS deviation from the mean:
442
443            PRMS = sqrt( sum( (x-mean(x))^2 )/(N-1)
444
445          The population deviation is the best-estimate of the deviation of
446          the population from which a sample is drawn.
447
448       ·  the median
449
450          The median is the 50th percentile data value.  Median is found by
451          medover, so WEIGHTING IS IGNORED FOR THE MEDIAN CALCULATION.
452
453       ·  the minimum
454
455       ·  the maximum
456
457       ·  the average absolute deviation:
458
459            AADEV = sum( abs(x-mean(x)) )/N
460
461       ·  RMS deviation from the mean:
462
463            RMS = sqrt(sum( (x-mean(x))^2 )/N)
464
465          (also known as the root-mean-square deviation, or the square root of
466          the variance)
467
468       This operator is a projection operator so the calculation will take
469       place over the final dimension. Thus if the input is N-dimensional each
470       returned value will be N-1 dimensional, to calculate the statistics for
471       the entire piddle either use "clump(-1)" directly on the piddle or call
472       "stats".
473
474       Bad values are simply ignored in the calculation, effectively reducing
475       the sample size.  If all data are bad then the output data are marked
476       bad.
477
478   stats
479       Calculates useful statistics on a piddle
480
481        ($mean,$prms,$median,$min,$max,$adev,$rms) = stats($piddle,[$weights]);
482
483       This utility calculates all the most useful quantities in one call.  It
484       works the same way as "statsover", except that the quantities are
485       calculated considering the entire input PDL as a single sample, rather
486       than as a collection of rows. See "statsover" for definitions of the
487       returned quantities.
488
489       Bad values are handled; if all input values are bad, then all of the
490       output values are flagged bad.
491
492   histogram
493         Signature: (in(n); int+[o] hist(m); double step; double min; int msize => m)
494
495       Calculates a histogram for given stepsize and minimum.
496
497        $h = histogram($data, $step, $min, $numbins);
498        $hist = zeroes $numbins;  # Put histogram in existing piddle.
499        histogram($data, $hist, $step, $min, $numbins);
500
501       The histogram will contain $numbins bins starting from $min, each $step
502       wide. The value in each bin is the number of values in $data that lie
503       within the bin limits.
504
505       Data below the lower limit is put in the first bin, and data above the
506       upper limit is put in the last bin.
507
508       The output is reset in a different threadloop so that you can take a
509       histogram of "$a(10,12)" into "$b(15)" and get the result you want.
510
511       For a higher-level interface, see hist.
512
513        pdl> p histogram(pdl(1,1,2),1,0,3)
514        [0 2 1]
515
516       histogram processes bad values.  It will set the bad-value flag of all
517       output piddles if the flag is set for any of the input piddles.
518
519   whistogram
520         Signature: (in(n); float+ wt(n);float+[o] hist(m); double step; double min; int msize => m)
521
522       Calculates a histogram from weighted data for given stepsize and
523       minimum.
524
525        $h = whistogram($data, $weights, $step, $min, $numbins);
526        $hist = zeroes $numbins;  # Put histogram in existing piddle.
527        whistogram($data, $weights, $hist, $step, $min, $numbins);
528
529       The histogram will contain $numbins bins starting from $min, each $step
530       wide. The value in each bin is the sum of the values in $weights that
531       correspond to values in $data that lie within the bin limits.
532
533       Data below the lower limit is put in the first bin, and data above the
534       upper limit is put in the last bin.
535
536       The output is reset in a different threadloop so that you can take a
537       histogram of "$a(10,12)" into "$b(15)" and get the result you want.
538
539        pdl> p whistogram(pdl(1,1,2), pdl(0.1,0.1,0.5), 1, 0, 4)
540        [0 0.2 0.5 0]
541
542       whistogram processes bad values.  It will set the bad-value flag of all
543       output piddles if the flag is set for any of the input piddles.
544
545   histogram2d
546         Signature: (ina(n); inb(n); int+[o] hist(ma,mb); double stepa; double mina; int masize => ma;
547                            double stepb; double minb; int mbsize => mb;)
548
549       Calculates a 2d histogram.
550
551        $h = histogram2d($datax, $datay, $stepx, $minx,
552              $nbinx, $stepy, $miny, $nbiny);
553        $hist = zeroes $nbinx, $nbiny;  # Put histogram in existing piddle.
554        histogram2d($datax, $datay, $hist, $stepx, $minx,
555              $nbinx, $stepy, $miny, $nbiny);
556
557       The histogram will contain $nbinx x $nbiny bins, with the lower limits
558       of the first one at "($minx, $miny)", and with bin size "($stepx,
559       $stepy)".  The value in each bin is the number of values in $datax and
560       $datay that lie within the bin limits.
561
562       Data below the lower limit is put in the first bin, and data above the
563       upper limit is put in the last bin.
564
565        pdl> p histogram2d(pdl(1,1,1,2,2),pdl(2,1,1,1,1),1,0,3,1,0,3)
566        [
567         [0 0 0]
568         [0 2 2]
569         [0 1 0]
570        ]
571
572       histogram2d processes bad values.  It will set the bad-value flag of
573       all output piddles if the flag is set for any of the input piddles.
574
575   whistogram2d
576         Signature: (ina(n); inb(n); float+ wt(n);float+[o] hist(ma,mb); double stepa; double mina; int masize => ma;
577                            double stepb; double minb; int mbsize => mb;)
578
579       Calculates a 2d histogram from weighted data.
580
581        $h = whistogram2d($datax, $datay, $weights,
582              $stepx, $minx, $nbinx, $stepy, $miny, $nbiny);
583        $hist = zeroes $nbinx, $nbiny;  # Put histogram in existing piddle.
584        whistogram2d($datax, $datay, $weights, $hist,
585              $stepx, $minx, $nbinx, $stepy, $miny, $nbiny);
586
587       The histogram will contain $nbinx x $nbiny bins, with the lower limits
588       of the first one at "($minx, $miny)", and with bin size "($stepx,
589       $stepy)".  The value in each bin is the sum of the values in $weights
590       that correspond to values in $datax and $datay that lie within the bin
591       limits.
592
593       Data below the lower limit is put in the first bin, and data above the
594       upper limit is put in the last bin.
595
596        pdl> p whistogram2d(pdl(1,1,1,2,2),pdl(2,1,1,1,1),pdl(0.1,0.2,0.3,0.4,0.5),1,0,3,1,0,3)
597        [
598         [  0   0   0]
599         [  0 0.5 0.9]
600         [  0 0.1   0]
601        ]
602
603       whistogram2d processes bad values.  It will set the bad-value flag of
604       all output piddles if the flag is set for any of the input piddles.
605
606   fibonacci
607         Signature: ([o]x(n))
608
609       Constructor - a vector with Fibonacci's sequence
610
611       fibonacci does not process bad values.  It will set the bad-value flag
612       of all output piddles if the flag is set for any of the input piddles.
613
614   append
615         Signature: (a(n); b(m); [o] c(mn))
616
617       append two piddles by concatenating along their first dimensions
618
619        $x = ones(2,4,7);
620        $y = sequence 5;
621        $c = $x->append($y);  # size of $c is now (7,4,7) (a jumbo-piddle ;)
622
623       "append" appends two piddles along their first dimensions. The rest of
624       the dimensions must be compatible in the threading sense. The resulting
625       size of the first dimension is the sum of the sizes of the first
626       dimensions of the two argument piddles - i.e. "n + m".
627
628       Similar functions include glue (below), which can append more than two
629       piddles along an arbitrary dimension, and cat, which can append more
630       than two piddles that all have the same sized dimensions.
631
632       append does not process bad values.  It will set the bad-value flag of
633       all output piddles if the flag is set for any of the input piddles.
634
635   glue
636         $c = $x->glue(<dim>,$y,...)
637
638       Glue two or more PDLs together along an arbitrary dimension (N-D
639       append).
640
641       Sticks $x, $y, and all following arguments together along the specified
642       dimension.  All other dimensions must be compatible in the threading
643       sense.
644
645       Glue is permissive, in the sense that every PDL is treated as having an
646       infinite number of trivial dimensions of order 1 -- so "$x->glue(3,$y)"
647       works, even if $x and $y are only one dimensional.
648
649       If one of the PDLs has no elements, it is ignored.  Likewise, if one of
650       them is actually the undefined value, it is treated as if it had no
651       elements.
652
653       If the first parameter is a defined perl scalar rather than a pdl, then
654       it is taken as a dimension along which to glue everything else, so you
655       can say "$cube = PDL::glue(3,@image_list);" if you like.
656
657       "glue" is implemented in pdl, using a combination of xchg and append.
658       It should probably be updated (one day) to a pure PP function.
659
660       Similar functions include append (above), which appends only two
661       piddles along their first dimension, and cat, which can append more
662       than two piddles that all have the same sized dimensions.
663
664   axisvalues
665         Signature: ([o,nc]a(n))
666
667       Internal routine
668
669       "axisvalues" is the internal primitive that implements axisvals and
670       alters its argument.
671
672       axisvalues does not process bad values.  It will set the bad-value flag
673       of all output piddles if the flag is set for any of the input piddles.
674
675   random
676       Constructor which returns piddle of random numbers
677
678        $x = random([type], $nx, $ny, $nz,...);
679        $x = random $y;
680
681       etc (see zeroes).
682
683       This is the uniform distribution between 0 and 1 (assumedly excluding 1
684       itself). The arguments are the same as "zeroes" (q.v.) - i.e. one can
685       specify dimensions, types or give a template.
686
687       You can use the perl function srand to seed the random generator. For
688       further details consult Perl's  srand documentation.
689
690   randsym
691       Constructor which returns piddle of random numbers
692
693        $x = randsym([type], $nx, $ny, $nz,...);
694        $x = randsym $y;
695
696       etc (see zeroes).
697
698       This is the uniform distribution between 0 and 1 (excluding both 0 and
699       1, cf random). The arguments are the same as "zeroes" (q.v.) - i.e. one
700       can specify dimensions, types or give a template.
701
702       You can use the perl function srand to seed the random generator. For
703       further details consult Perl's  srand documentation.
704
705   grandom
706       Constructor which returns piddle of Gaussian random numbers
707
708        $x = grandom([type], $nx, $ny, $nz,...);
709        $x = grandom $y;
710
711       etc (see zeroes).
712
713       This is generated using the math library routine "ndtri".
714
715       Mean = 0, Stddev = 1
716
717       You can use the perl function srand to seed the random generator. For
718       further details consult Perl's  srand documentation.
719
720   vsearch
721         Signature: ( vals(); xs(n); [o] indx(); [\%options] )
722
723       Efficiently search for values in a sorted piddle, returning indices.
724
725         $idx = vsearch( $vals, $x, [\%options] );
726         vsearch( $vals, $x, $idx, [\%options ] );
727
728       vsearch performs a binary search in the ordered piddle $x, for the
729       values from $vals piddle, returning indices into $x.  What is a
730       "match", and the meaning of the returned indices, are determined by the
731       options.
732
733       The "mode" option indicates which method of searching to use, and may
734       be one of:
735
736       "sample"
737           invoke vsearch_sample, returning indices appropriate for sampling
738           within a distribution.
739
740       "insert_leftmost"
741           invoke vsearch_insert_leftmost, returning the left-most possible
742           insertion point which still leaves the piddle sorted.
743
744       "insert_rightmost"
745           invoke vsearch_insert_rightmost, returning the right-most possible
746           insertion point which still leaves the piddle sorted.
747
748       "match"
749           invoke vsearch_match, returning the index of a matching element,
750           else -(insertion point + 1)
751
752       "bin_inclusive"
753           invoke vsearch_bin_inclusive, returning an index appropriate for
754           binning on a grid where the left bin edges are inclusive of the
755           bin. See below for further explanation of the bin.
756
757       "bin_exclusive"
758           invoke vsearch_bin_exclusive, returning an index appropriate for
759           binning on a grid where the left bin edges are exclusive of the
760           bin. See below for further explanation of the bin.
761
762       The default value of "mode" is "sample".
763
764         use PDL;
765
766         my @modes = qw( sample insert_leftmost insert_rightmost match
767                         bin_inclusive bin_exclusive );
768
769         # Generate a sequence of 3 zeros, 3 ones, ..., 3 fours.
770         my $x = zeroes(3,5)->yvals->flat;
771
772         for my $mode ( @modes ) {
773           # if the value is in $x
774           my $contained = 2;
775           my $idx_contained = vsearch( $contained, $x, { mode => $mode } );
776           my $x_contained = $x->copy;
777           $x_contained->slice( $idx_contained ) .= 9;
778
779           # if the value is not in $x
780           my $not_contained = 1.5;
781           my $idx_not_contained = vsearch( $not_contained, $x, { mode => $mode } );
782           my $x_not_contained = $x->copy;
783           $x_not_contained->slice( $idx_not_contained ) .= 9;
784
785           print sprintf("%-23s%30s\n", '$x', $x);
786           print sprintf("%-23s%30s\n",   "$mode ($contained)", $x_contained);
787           print sprintf("%-23s%30s\n\n", "$mode ($not_contained)", $x_not_contained);
788         }
789
790         # $x                     [0 0 0 1 1 1 2 2 2 3 3 3 4 4 4]
791         # sample (2)             [0 0 0 1 1 1 9 2 2 3 3 3 4 4 4]
792         # sample (1.5)           [0 0 0 1 1 1 9 2 2 3 3 3 4 4 4]
793         #
794         # $x                     [0 0 0 1 1 1 2 2 2 3 3 3 4 4 4]
795         # insert_leftmost (2)    [0 0 0 1 1 1 9 2 2 3 3 3 4 4 4]
796         # insert_leftmost (1.5)  [0 0 0 1 1 1 9 2 2 3 3 3 4 4 4]
797         #
798         # $x                     [0 0 0 1 1 1 2 2 2 3 3 3 4 4 4]
799         # insert_rightmost (2)   [0 0 0 1 1 1 2 2 2 9 3 3 4 4 4]
800         # insert_rightmost (1.5) [0 0 0 1 1 1 9 2 2 3 3 3 4 4 4]
801         #
802         # $x                     [0 0 0 1 1 1 2 2 2 3 3 3 4 4 4]
803         # match (2)              [0 0 0 1 1 1 2 9 2 3 3 3 4 4 4]
804         # match (1.5)            [0 0 0 1 1 1 2 2 9 3 3 3 4 4 4]
805         #
806         # $x                     [0 0 0 1 1 1 2 2 2 3 3 3 4 4 4]
807         # bin_inclusive (2)      [0 0 0 1 1 1 2 2 9 3 3 3 4 4 4]
808         # bin_inclusive (1.5)    [0 0 0 1 1 9 2 2 2 3 3 3 4 4 4]
809         #
810         # $x                     [0 0 0 1 1 1 2 2 2 3 3 3 4 4 4]
811         # bin_exclusive (2)      [0 0 0 1 1 9 2 2 2 3 3 3 4 4 4]
812         # bin_exclusive (1.5)    [0 0 0 1 1 9 2 2 2 3 3 3 4 4 4]
813
814       Also see vsearch_sample, vsearch_insert_leftmost,
815       vsearch_insert_rightmost, vsearch_match, vsearch_bin_inclusive, and
816       vsearch_bin_exclusive
817
818   vsearch_sample
819         Signature: (vals(); x(n); indx [o]idx())
820
821       Search for values in a sorted array, return index appropriate for
822       sampling from a distribution
823
824         $idx = vsearch_sample($vals, $x);
825
826       $x must be sorted, but may be in decreasing or increasing order.
827
828       vsearch_sample returns an index I for each value V of $vals appropriate
829       for sampling $vals
830
831       I has the following properties:
832
833       ·   if $x is sorted in increasing order
834
835                     V <= x[0]  : I = 0
836             x[0]  < V <= x[-1] : I s.t. x[I-1] < V <= x[I]
837             x[-1] < V          : I = $x->nelem -1
838
839       ·   if $x is sorted in decreasing order
840
841                      V > x[0]  : I = 0
842             x[0]  >= V > x[-1] : I s.t. x[I] >= V > x[I+1]
843             x[-1] >= V         : I = $x->nelem - 1
844
845       If all elements of $x are equal, I = $x->nelem - 1.
846
847       If $x contains duplicated elements, I is the index of the leftmost (by
848       position in array) duplicate if V matches.
849
850       This function is useful e.g. when you have a list of probabilities for
851       events and want to generate indices to events:
852
853        $x = pdl(.01,.86,.93,1); # Barnsley IFS probabilities cumulatively
854        $y = random 20;
855        $c = vsearch_sample($y, $x); # Now, $c will have the appropriate distr.
856
857       It is possible to use the cumusumover function to obtain cumulative
858       probabilities from absolute probabilities.
859
860       needs major (?) work to handles bad values
861
862   vsearch_insert_leftmost
863         Signature: (vals(); x(n); indx [o]idx())
864
865       Determine the insertion point for values in a sorted array, inserting
866       before duplicates.
867
868         $idx = vsearch_insert_leftmost($vals, $x);
869
870       $x must be sorted, but may be in decreasing or increasing order.
871
872       vsearch_insert_leftmost returns an index I for each value V of $vals
873       equal to the leftmost position (by index in array) within $x that V may
874       be inserted and still maintain the order in $x.
875
876       Insertion at index I involves shifting elements I and higher of $x to
877       the right by one and setting the now empty element at index I to V.
878
879       I has the following properties:
880
881       ·   if $x is sorted in increasing order
882
883                     V <= x[0]  : I = 0
884             x[0]  < V <= x[-1] : I s.t. x[I-1] < V <= x[I]
885             x[-1] < V          : I = $x->nelem
886
887       ·   if $x is sorted in decreasing order
888
889                      V >  x[0]  : I = -1
890             x[0]  >= V >= x[-1] : I s.t. x[I] >= V > x[I+1]
891             x[-1] >= V          : I = $x->nelem -1
892
893       If all elements of $x are equal,
894
895           i = 0
896
897       If $x contains duplicated elements, I is the index of the leftmost (by
898       index in array) duplicate if V matches.
899
900       needs major (?) work to handles bad values
901
902   vsearch_insert_rightmost
903         Signature: (vals(); x(n); indx [o]idx())
904
905       Determine the insertion point for values in a sorted array, inserting
906       after duplicates.
907
908         $idx = vsearch_insert_rightmost($vals, $x);
909
910       $x must be sorted, but may be in decreasing or increasing order.
911
912       vsearch_insert_rightmost returns an index I for each value V of $vals
913       equal to the rightmost position (by index in array) within $x that V
914       may be inserted and still maintain the order in $x.
915
916       Insertion at index I involves shifting elements I and higher of $x to
917       the right by one and setting the now empty element at index I to V.
918
919       I has the following properties:
920
921       ·   if $x is sorted in increasing order
922
923                      V < x[0]  : I = 0
924             x[0]  <= V < x[-1] : I s.t. x[I-1] <= V < x[I]
925             x[-1] <= V         : I = $x->nelem
926
927       ·   if $x is sorted in decreasing order
928
929                     V >= x[0]  : I = -1
930             x[0]  > V >= x[-1] : I s.t. x[I] >= V > x[I+1]
931             x[-1] > V          : I = $x->nelem -1
932
933       If all elements of $x are equal,
934
935           i = $x->nelem - 1
936
937       If $x contains duplicated elements, I is the index of the leftmost (by
938       index in array) duplicate if V matches.
939
940       needs major (?) work to handles bad values
941
942   vsearch_match
943         Signature: (vals(); x(n); indx [o]idx())
944
945       Match values against a sorted array.
946
947         $idx = vsearch_match($vals, $x);
948
949       $x must be sorted, but may be in decreasing or increasing order.
950
951       vsearch_match returns an index I for each value V of $vals.  If V
952       matches an element in $x, I is the index of that element, otherwise it
953       is -( insertion_point + 1 ), where insertion_point is an index in $x
954       where V may be inserted while maintaining the order in $x.  If $x has
955       duplicated values, I may refer to any of them.
956
957       needs major (?) work to handles bad values
958
959   vsearch_bin_inclusive
960         Signature: (vals(); x(n); indx [o]idx())
961
962       Determine the index for values in a sorted array of bins, lower bound
963       inclusive.
964
965         $idx = vsearch_bin_inclusive($vals, $x);
966
967       $x must be sorted, but may be in decreasing or increasing order.
968
969       $x represents the edges of contiguous bins, with the first and last
970       elements representing the outer edges of the outer bins, and the inner
971       elements the shared bin edges.
972
973       The lower bound of a bin is inclusive to the bin, its outer bound is
974       exclusive to it.  vsearch_bin_inclusive returns an index I for each
975       value V of $vals
976
977       I has the following properties:
978
979       ·   if $x is sorted in increasing order
980
981                      V < x[0]  : I = -1
982             x[0]  <= V < x[-1] : I s.t. x[I] <= V < x[I+1]
983             x[-1] <= V         : I = $x->nelem - 1
984
985       ·   if $x is sorted in decreasing order
986
987                      V >= x[0]  : I = 0
988             x[0]  >  V >= x[-1] : I s.t. x[I+1] > V >= x[I]
989             x[-1] >  V          : I = $x->nelem
990
991       If all elements of $x are equal,
992
993           i = $x->nelem - 1
994
995       If $x contains duplicated elements, I is the index of the righmost (by
996       index in array) duplicate if V matches.
997
998       needs major (?) work to handles bad values
999
1000   vsearch_bin_exclusive
1001         Signature: (vals(); x(n); indx [o]idx())
1002
1003       Determine the index for values in a sorted array of bins, lower bound
1004       exclusive.
1005
1006         $idx = vsearch_bin_exclusive($vals, $x);
1007
1008       $x must be sorted, but may be in decreasing or increasing order.
1009
1010       $x represents the edges of contiguous bins, with the first and last
1011       elements representing the outer edges of the outer bins, and the inner
1012       elements the shared bin edges.
1013
1014       The lower bound of a bin is exclusive to the bin, its upper bound is
1015       inclusive to it.  vsearch_bin_exclusive returns an index I for each
1016       value V of $vals.
1017
1018       I has the following properties:
1019
1020       ·   if $x is sorted in increasing order
1021
1022                      V <= x[0]  : I = -1
1023             x[0]  <  V <= x[-1] : I s.t. x[I] < V <= x[I+1]
1024             x[-1] <  V          : I = $x->nelem - 1
1025
1026       ·   if $x is sorted in decreasing order
1027
1028                      V >  x[0]  : I = 0
1029             x[0]  >= V >  x[-1] : I s.t. x[I-1] >= V > x[I]
1030             x[-1] >= V          : I = $x->nelem
1031
1032       If all elements of $x are equal,
1033
1034           i = $x->nelem - 1
1035
1036       If $x contains duplicated elements, I is the index of the righmost (by
1037       index in array) duplicate if V matches.
1038
1039       needs major (?) work to handles bad values
1040
1041   interpolate
1042         Signature: (xi(); x(n); y(n); [o] yi(); int [o] err())
1043
1044       routine for 1D linear interpolation
1045
1046        ( $yi, $err ) = interpolate($xi, $x, $y)
1047
1048       Given a set of points "($x,$y)", use linear interpolation to find the
1049       values $yi at a set of points $xi.
1050
1051       "interpolate" uses a binary search to find the suspects, er...,
1052       interpolation indices and therefore abscissas (ie $x) have to be
1053       strictly ordered (increasing or decreasing).  For interpolation at lots
1054       of closely spaced abscissas an approach that uses the last index found
1055       as a start for the next search can be faster (compare Numerical Recipes
1056       "hunt" routine). Feel free to implement that on top of the binary
1057       search if you like. For out of bounds values it just does a linear
1058       extrapolation and sets the corresponding element of $err to 1, which is
1059       otherwise 0.
1060
1061       See also interpol, which uses the same routine, differing only in the
1062       handling of extrapolation - an error message is printed rather than
1063       returning an error piddle.
1064
1065       needs major (?) work to handles bad values
1066
1067   interpol
1068        Signature: (xi(); x(n); y(n); [o] yi())
1069
1070       routine for 1D linear interpolation
1071
1072        $yi = interpol($xi, $x, $y)
1073
1074       "interpol" uses the same search method as interpolate, hence $x must be
1075       strictly ordered (either increasing or decreasing).  The difference
1076       occurs in the handling of out-of-bounds values; here an error message
1077       is printed.
1078
1079   interpND
1080       Interpolate values from an N-D piddle, with switchable method
1081
1082         $source = 10*xvals(10,10) + yvals(10,10);
1083         $index = pdl([[2.2,3.5],[4.1,5.0]],[[6.0,7.4],[8,9]]);
1084         print $source->interpND( $index );
1085
1086       InterpND acts like indexND, collapsing $index by lookup into $source;
1087       but it does interpolation rather than direct sampling.  The
1088       interpolation method and boundary condition are switchable via an
1089       options hash.
1090
1091       By default, linear or sample interpolation is used, with constant value
1092       outside the boundaries of the source pdl.  No dataflow occurs, because
1093       in general the output is computed rather than indexed.
1094
1095       All the interpolation methods treat the pixels as value-centered, so
1096       the "sample" method will return "$a->(0)" for coordinate values on the
1097       set [-0.5,0.5), and all methods will return "$a->(1)" for a coordinate
1098       value of exactly 1.
1099
1100       Recognized options:
1101
1102       method
1103          Values can be:
1104
1105          ·  0, s, sample, Sample (default for integer source types)
1106
1107             The nearest value is taken. Pixels are regarded as centered on
1108             their respective integer coordinates (no offset from the linear
1109             case).
1110
1111          ·  1, l, linear, Linear (default for floating point source types)
1112
1113             The values are N-linearly interpolated from an N-dimensional cube
1114             of size 2.
1115
1116          ·  3, c, cube, cubic, Cubic
1117
1118             The values are interpolated using a local cubic fit to the data.
1119             The fit is constrained to match the original data and its
1120             derivative at the data points.  The second derivative of the fit
1121             is not continuous at the data points.  Multidimensional datasets
1122             are interpolated by the successive-collapse method.
1123
1124             (Note that the constraint on the first derivative causes a small
1125             amount of ringing around sudden features such as step functions).
1126
1127          ·  f, fft, fourier, Fourier
1128
1129             The source is Fourier transformed, and the interpolated values
1130             are explicitly calculated from the coefficients.  The boundary
1131             condition option is ignored -- periodic boundaries are imposed.
1132
1133             If you pass in the option "fft", and it is a list (ARRAY) ref,
1134             then it is a stash for the magnitude and phase of the source FFT.
1135             If the list has two elements then they are taken as already
1136             computed; otherwise they are calculated and put in the stash.
1137
1138       b, bound, boundary, Boundary
1139          This option is passed unmodified into indexND, which is used as the
1140          indexing engine for the interpolation.  Some current allowed values
1141          are 'extend', 'periodic', 'truncate', and 'mirror' (default is
1142          'truncate').
1143
1144       bad
1145          contains the fill value used for 'truncate' boundary.  (default 0)
1146
1147       fft
1148          An array ref whose associated list is used to stash the FFT of the
1149          source data, for the FFT method.
1150
1151   one2nd
1152       Converts a one dimensional index piddle to a set of ND coordinates
1153
1154        @coords=one2nd($x, $indices)
1155
1156       returns an array of piddles containing the ND indexes corresponding to
1157       the one dimensional list indices. The indices are assumed to correspond
1158       to array $x clumped using "clump(-1)". This routine is used in the old
1159       vector form of whichND, but is useful on its own occasionally.
1160
1161       Returned piddles have the indx datatype.  $indices can have values
1162       larger than "$x->nelem" but negative values in $indices will not give
1163       the answer you expect.
1164
1165        pdl> $x=pdl [[[1,2],[-1,1]], [[0,-3],[3,2]]]; $c=$x->clump(-1)
1166        pdl> $maxind=maximum_ind($c); p $maxind;
1167        6
1168        pdl> print one2nd($x, maximum_ind($c))
1169        0 1 1
1170        pdl> p $x->at(0,1,1)
1171        3
1172
1173   which
1174         Signature: (mask(n); indx [o] inds(m))
1175
1176       Returns indices of non-zero values from a 1-D PDL
1177
1178        $i = which($mask);
1179
1180       returns a pdl with indices for all those elements that are nonzero in
1181       the mask. Note that the returned indices will be 1D. If you feed in a
1182       multidimensional mask, it will be flattened before the indices are
1183       calculated.  See also whichND for multidimensional masks.
1184
1185       If you want to index into the original mask or a similar piddle with
1186       output from "which", remember to flatten it before calling index:
1187
1188         $data = random 5, 5;
1189         $idx = which $data > 0.5; # $idx is now 1D
1190         $bigsum = $data->flat->index($idx)->sum;  # flatten before indexing
1191
1192       Compare also where for similar functionality.
1193
1194       SEE ALSO:
1195
1196       which_both returns separately the indices of both zero and nonzero
1197       values in the mask.
1198
1199       where returns associated values from a data PDL, rather than indices
1200       into the mask PDL.
1201
1202       whichND returns N-D indices into a multidimensional PDL.
1203
1204        pdl> $x = sequence(10); p $x
1205        [0 1 2 3 4 5 6 7 8 9]
1206        pdl> $indx = which($x>6); p $indx
1207        [7 8 9]
1208
1209       which processes bad values.  It will set the bad-value flag of all
1210       output piddles if the flag is set for any of the input piddles.
1211
1212   which_both
1213         Signature: (mask(n); indx [o] inds(m); indx [o]notinds(q))
1214
1215       Returns indices of zero and nonzero values in a mask PDL
1216
1217        ($i, $c_i) = which_both($mask);
1218
1219       This works just as which, but the complement of $i will be in $c_i.
1220
1221        pdl> $x = sequence(10); p $x
1222        [0 1 2 3 4 5 6 7 8 9]
1223        pdl> ($small, $big) = which_both ($x >= 5); p "$small\n $big"
1224        [5 6 7 8 9]
1225        [0 1 2 3 4]
1226
1227       which_both processes bad values.  It will set the bad-value flag of all
1228       output piddles if the flag is set for any of the input piddles.
1229
1230   where
1231       Use a mask to select values from one or more data PDLs
1232
1233       "where" accepts one or more data piddles and a mask piddle.  It returns
1234       a list of output piddles, corresponding to the input data piddles.
1235       Each output piddle is a 1-dimensional list of values in its
1236       corresponding data piddle. The values are drawn from locations where
1237       the mask is nonzero.
1238
1239       The output PDLs are still connected to the original data PDLs, for the
1240       purpose of dataflow.
1241
1242       "where" combines the functionality of which and index into a single
1243       operation.
1244
1245       BUGS:
1246
1247       While "where" works OK for most N-dimensional cases, it does not thread
1248       properly over (for example) the (N+1)th dimension in data that is
1249       compared to an N-dimensional mask.  Use "whereND" for that.
1250
1251        $i = $x->where($x+5 > 0); # $i contains those elements of $x
1252                                  # where mask ($x+5 > 0) is 1
1253        $i .= -5;  # Set those elements (of $x) to -5. Together, these
1254                   # commands clamp $x to a maximum of -5.
1255
1256       It is also possible to use the same mask for several piddles with the
1257       same call:
1258
1259        ($i,$j,$k) = where($x,$y,$z, $x+5>0);
1260
1261       Note: $i is always 1-D, even if $x is >1-D.
1262
1263       WARNING: The first argument (the values) and the second argument (the
1264       mask) currently have to have the exact same dimensions (or horrible
1265       things happen). You *cannot* thread over a smaller mask, for example.
1266
1267   whereND
1268       "where" with support for ND masks and threading
1269
1270       "whereND" accepts one or more data piddles and a mask piddle.  It
1271       returns a list of output piddles, corresponding to the input data
1272       piddles.  The values are drawn from locations where the mask is
1273       nonzero.
1274
1275       "whereND" differs from "where" in that the mask dimensionality is
1276       preserved which allows for proper threading of the selection operation
1277       over higher dimensions.
1278
1279       As with "where" the output PDLs are still connected to the original
1280       data PDLs, for the purpose of dataflow.
1281
1282         $sdata = whereND $data, $mask
1283         ($s1, $s2, ..., $sn) = whereND $d1, $d2, ..., $dn, $mask
1284
1285         where
1286
1287           $data is M dimensional
1288           $mask is N < M dimensional
1289           dims($data) 1..N == dims($mask) 1..N
1290           with threading over N+1 to M dimensions
1291
1292         $data   = sequence(4,3,2);   # example data array
1293         $mask4  = (random(4)>0.5);   # example 1-D mask array, has $n4 true values
1294         $mask43 = (random(4,3)>0.5); # example 2-D mask array, has $n43 true values
1295         $sdat4  = whereND $data, $mask4;   # $sdat4 is a [$n4,3,2] pdl
1296         $sdat43 = whereND $data, $mask43;  # $sdat43 is a [$n43,2] pdl
1297
1298       Just as with "where", you can use the returned value in an assignment.
1299       That means that both of these examples are valid:
1300
1301         # Used to create a new slice stored in $sdat4:
1302         $sdat4 = $data->whereND($mask4);
1303         $sdat4 .= 0;
1304         # Used in lvalue context:
1305         $data->whereND($mask4) .= 0;
1306
1307   whichND
1308       Return the coordinates of non-zero values in a mask.
1309
1310       WhichND returns the N-dimensional coordinates of each nonzero value in
1311       a mask PDL with any number of dimensions.  The returned values arrive
1312       as an array-of-vectors suitable for use in indexND or range.
1313
1314        $coords = whichND($mask);
1315
1316       returns a PDL containing the coordinates of the elements that are non-
1317       zero in $mask, suitable for use in indexND.  The 0th dimension contains
1318       the full coordinate listing of each point; the 1st dimension lists all
1319       the points.  For example, if $mask has rank 4 and 100 matching
1320       elements, then $coords has dimension 4x100.
1321
1322       If no such elements exist, then whichND returns a structured empty PDL:
1323       an Nx0 PDL that contains no values (but matches, threading-wise, with
1324       the vectors that would be produced if such elements existed).
1325
1326       DEPRECATED BEHAVIOR IN LIST CONTEXT:
1327
1328       whichND once delivered different values in list context than in scalar
1329       context, for historical reasons.  In list context, it returned the
1330       coordinates transposed, as a collection of 1-PDLs (one per dimension)
1331       in a list.  This usage is deprecated in PDL 2.4.10, and will cause a
1332       warning to be issued every time it is encountered.  To avoid the
1333       warning, you can set the global variable "$PDL::whichND" to 's' to get
1334       scalar behavior in all contexts, or to 'l' to get list behavior in list
1335       context.
1336
1337       In later versions of PDL, the deprecated behavior will disappear.
1338       Deprecated list context whichND expressions can be replaced with:
1339
1340           @list = $x->whichND->mv(0,-1)->dog;
1341
1342       SEE ALSO:
1343
1344       which finds coordinates of nonzero values in a 1-D mask.
1345
1346       where extracts values from a data PDL that are associated with nonzero
1347       values in a mask PDL.
1348
1349        pdl> $s=sequence(10,10,3,4)
1350        pdl> ($x, $y, $z, $w)=whichND($s == 203); p $x, $y, $z, $w
1351        [3] [0] [2] [0]
1352        pdl> print $s->at(list(cat($x,$y,$z,$w)))
1353        203
1354
1355   setops
1356       Implements simple set operations like union and intersection
1357
1358          Usage: $set = setops($x, <OPERATOR>, $y);
1359
1360       The operator can be "OR", "XOR" or "AND". This is then applied to $x
1361       viewed as a set and $y viewed as a set. Set theory says that a set may
1362       not have two or more identical elements, but setops takes care of this
1363       for you, so "$x=pdl(1,1,2)" is OK. The functioning is as follows:
1364
1365       "OR"
1366           The resulting vector will contain the elements that are either in
1367           $x or in $y or both. This is the union in set operation terms
1368
1369       "XOR"
1370           The resulting vector will contain the elements that are either in
1371           $x or $y, but not in both. This is
1372
1373                Union($x, $y) - Intersection($x, $y)
1374
1375           in set operation terms.
1376
1377       "AND"
1378           The resulting vector will contain the intersection of $x and $y, so
1379           the elements that are in both $x and $y. Note that for convenience
1380           this operation is also aliased to intersect.
1381
1382       It should be emphasized that these routines are used when one or both
1383       of the sets $x, $y are hard to calculate or that you get from a
1384       separate subroutine.
1385
1386       Finally IDL users might be familiar with Craig Markwardt's
1387       "cmset_op.pro" routine which has inspired this routine although it was
1388       written independently However the present routine has a few less
1389       options (but see the examples)
1390
1391       You will very often use these functions on an index vector, so that is
1392       what we will show here. We will in fact something slightly silly. First
1393       we will find all squares that are also cubes below 10000.
1394
1395       Create a sequence vector:
1396
1397         pdl> $x = sequence(10000)
1398
1399       Find all odd and even elements:
1400
1401         pdl> ($even, $odd) = which_both( ($x % 2) == 0)
1402
1403       Find all squares
1404
1405         pdl> $squares= which(ceil(sqrt($x)) == floor(sqrt($x)))
1406
1407       Find all cubes (being careful with roundoff error!)
1408
1409         pdl> $cubes= which(ceil($x**(1.0/3.0)) == floor($x**(1.0/3.0)+1e-6))
1410
1411       Then find all squares that are cubes:
1412
1413         pdl> $both = setops($squares, 'AND', $cubes)
1414
1415       And print these (assumes that "PDL::NiceSlice" is loaded!)
1416
1417         pdl> p $x($both)
1418          [0 1 64 729 4096]
1419
1420       Then find all numbers that are either cubes or squares, but not both:
1421
1422         pdl> $cube_xor_square = setops($squares, 'XOR', $cubes)
1423
1424         pdl> p $cube_xor_square->nelem()
1425          112
1426
1427       So there are a total of 112 of these!
1428
1429       Finally find all odd squares:
1430
1431         pdl> $odd_squares = setops($squares, 'AND', $odd)
1432
1433       Another common occurrence is to want to get all objects that are in $x
1434       and in the complement of $y. But it is almost always best to create the
1435       complement explicitly since the universe that both are taken from is
1436       not known. Thus use which_both if possible to keep track of
1437       complements.
1438
1439       If this is impossible the best approach is to make a temporary:
1440
1441       This creates an index vector the size of the universe of the sets and
1442       set all elements in $y to 0
1443
1444         pdl> $tmp = ones($n_universe); $tmp($y) .= 0;
1445
1446       This then finds the complement of $y
1447
1448         pdl> $C_b = which($tmp == 1);
1449
1450       and this does the final selection:
1451
1452         pdl> $set = setops($x, 'AND', $C_b)
1453
1454   intersect
1455       Calculate the intersection of two piddles
1456
1457          Usage: $set = intersect($x, $y);
1458
1459       This routine is merely a simple interface to setops. See that for more
1460       information
1461
1462       Find all numbers less that 100 that are of the form 2*y and 3*x
1463
1464        pdl> $x=sequence(100)
1465        pdl> $factor2 = which( ($x % 2) == 0)
1466        pdl> $factor3 = which( ($x % 3) == 0)
1467        pdl> $ii=intersect($factor2, $factor3)
1468        pdl> p $x($ii)
1469        [0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96]
1470

AUTHOR

1472       Copyright (C) Tuomas J. Lukka 1997 (lukka@husc.harvard.edu).
1473       Contributions by Christian Soeller (c.soeller@auckland.ac.nz), Karl
1474       Glazebrook (kgb@aaoepp.aao.gov.au), Craig DeForest
1475       (deforest@boulder.swri.edu) and Jarle Brinchmann (jarle@astro.up.pt)
1476       All rights reserved. There is no warranty. You are allowed to
1477       redistribute this software / documentation under certain conditions.
1478       For details, see the file COPYING in the PDL distribution. If this file
1479       is separated from the PDL distribution, the copyright notice should be
1480       included in the file.
1481
1482       Updated for CPAN viewing compatibility by David Mertens.
1483
1484
1485
1486perl v5.32.0                      2020-09-17                      Primitive(3)
Impressum