This file is indexed.

/usr/include/vigra/sampling.hxx is in libvigraimpex-dev 1.10.0+git20160211.167be93+dfsg-5ubuntu1.

This file is owned by root:root, with mode 0o644.

The actual contents of the file can be viewed below.

  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
/************************************************************************/
/*                                                                      */
/*        Copyright 2008-2009 by  Ullrich Koethe and Rahul Nair         */
/*                                                                      */
/*    This file is part of the VIGRA computer vision library.           */
/*    The VIGRA Website is                                              */
/*        http://hci.iwr.uni-heidelberg.de/vigra/                       */
/*    Please direct questions, bug reports, and contributions to        */
/*        ullrich.koethe@iwr.uni-heidelberg.de    or                    */
/*        vigra@informatik.uni-hamburg.de                               */
/*                                                                      */
/*    Permission is hereby granted, free of charge, to any person       */
/*    obtaining a copy of this software and associated documentation    */
/*    files (the "Software"), to deal in the Software without           */
/*    restriction, including without limitation the rights to use,      */
/*    copy, modify, merge, publish, distribute, sublicense, and/or      */
/*    sell copies of the Software, and to permit persons to whom the    */
/*    Software is furnished to do so, subject to the following          */
/*    conditions:                                                       */
/*                                                                      */
/*    The above copyright notice and this permission notice shall be    */
/*    included in all copies or substantial portions of the             */
/*    Software.                                                         */
/*                                                                      */
/*    THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND    */
/*    EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES   */
/*    OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND          */
/*    NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT       */
/*    HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,      */
/*    WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING      */
/*    FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR     */
/*    OTHER DEALINGS IN THE SOFTWARE.                                   */
/*                                                                      */
/************************************************************************/

#ifndef VIGRA_SAMPLING_HXX
#define VIGRA_SAMPLING_HXX

#include "array_vector.hxx"
#include "random.hxx"
#include <map>
#include <memory>
#include <cmath>

namespace vigra
{

/** \addtogroup MachineLearning Machine Learning
**/
//@{


/**\brief Options object for the Sampler class.
 
  <b>usage:</b>
 
  \code
  SamplerOptions opt =  SamplerOptions()
                               .withReplacement()
                               .sampleProportion(0.5);
  \endcode
 
  Note that the return value of all methods is <tt>*this</tt> which makes
  concatenating of options as above possible.
*/
class SamplerOptions
{
  public:

    double sample_proportion;
    unsigned int sample_size;
    bool   sample_with_replacement;
    bool   stratified_sampling;
    
    SamplerOptions()
    : sample_proportion(1.0),
      sample_size(0), 
      sample_with_replacement(true),
      stratified_sampling(false)
    {}

        /**\brief Sample from training population with replacement.
         *
         * <br> Default: true
         */
    SamplerOptions& withReplacement(bool in = true)
    {
        sample_with_replacement = in;
        return *this;
    }

        /**\brief Sample from training population without replacement.
         *
         * <br> Default (if you don't call this function): false
         */
    SamplerOptions& withoutReplacement(bool in = true)
    {
        sample_with_replacement = !in;
        return *this;
    }

        /**\brief Draw the given number of samples.
         * If stratifiedSampling is true, the \a size is equally distributed
         * across all strata (e.g. <tt>size / strataCount</tt> samples are taken 
         * from each stratum, subject to rounding).
         *
         * <br> Default: 0 (i.e. determine the count by means of sampleProportion())
         */
    SamplerOptions& sampleSize(unsigned int size)
    {
        sample_size = size;
        return *this;
    }


        /**\brief Determine the number of samples to draw as a proportion of the total
         * number. That is, we draw <tt>count = totalCount * proportion</tt> samples. 
         * This option is overridden when an absolute count is specified by sampleSize().
         * 
         * If stratifiedSampling is true, the count is equally distributed
         * across all strata (e.g. <tt>totalCount * proportion / strataCount</tt> samples are taken 
         * from each stratum).
         *
         * <br> Default: 1.0
         */
    SamplerOptions& sampleProportion(double proportion)
    {
        vigra_precondition(proportion >= 0.0,
               "SamplerOptions::sampleProportion(): argument must not be negative.");
        sample_proportion = proportion;
        return *this;
    }

        /**\brief Draw equally many samples from each "stratum". 
         *  A stratum is a group of like entities, e.g. pixels belonging 
         *  to the same object class. This is useful to create balanced samples  
         *  when the class probabilities are very unbalanced (e.g.
         *  when there are many background and few foreground pixels).
         *  Stratified sampling thus avoids that a trained classifier is biased 
         *  towards the majority class. 
         *
         * <br> Default (if you don't call this function): false
         */
    SamplerOptions& stratified(bool in = true)
    {
        stratified_sampling = in;
        return *this;
    }
};

/************************************************************/
/*                                                          */
/*                        Sampler                           */
/*                                                          */
/************************************************************/

/** \brief Create random samples from a sequence of indices.

    Selecting data items at random is a basic task of machine learning,
    for example in boostrapping, RandomForest training, and cross validation.
    This class implements various ways to select random samples via their indices. 
    Indices are assumed to be consecutive in
    the range <tt>0 &lt;= index &lt; total_sample_count</tt>.
    
    The class always contains a current sample which can be accessed by 
    the index operator or by the function sampledIndices(). The indices
    that are not in the current sample (out-of-bag indices) can be accessed
    via the function oobIndices().
    
    The sampling method (with/without replacement, stratified or not) and the
    number of samples to draw are determined by the option object 
    SamplerOptions.
    
    <b>Usage:</b>
    
    <b>\#include</b> \<vigra/sampling.hxx\><br>
    Namespace: vigra
    
    Create a Sampler with default options, i.e. sample as many indices as there 
    are data elements, with replacement. On average, the sample will contain 
    <tt>0.63*totalCount</tt> distinct indices.
    
    \code
    int totalCount = 10000;   // total number of data elements
    int numberOfSamples = 20; // repeat experiment 20 times 
    Sampler<> sampler(totalCount);
    for(int k=0; k<numberOfSamples; ++k)
    {
        // process current sample
        for(int i=0; i<sampler.sampleSize(); ++i)
        {
            int currentIndex = sampler[i];
            processData(data[currentIndex]);
        }
        // create next sample
        sampler.sample();
    }
    \endcode
    
    Create a Sampler for stratified sampling, without replacement.
    
    \code
    // prepare the strata (i.e. specify which stratum each element belongs to)
    int stratumSize1 = 2000, stratumSize2 = 8000,
        totalCount = stratumSize1 + stratumSize2;
    ArrayVerctor<int> strata(totalCount);
    for(int i=0; i<stratumSize1; ++i)
        strata[i] = 1;
    for(int i=stratumSize1; i<stratumSize2; ++i)
        strata[i] = 2;
        
    int sampleSize = 200; // i.e. sample 100 elements from each of the two strata
    int numberOfSamples = 20; // repeat experiment 20 times 
    Sampler<> stratifiedSampler(strata.begin(), strata.end(),
                     SamplerOptions().withoutReplacement().stratified().sampleSize(sampleSize));
    // create first sample
    sampler.sample();

    for(int k=0; k<numberOfSamples; ++k)
    {
        // process current sample
        for(int i=0; i<sampler.sampleSize(); ++i)
        {
            int currentIndex = sampler[i];
            processData(data[currentIndex]);
        }
        // create next sample
        sampler.sample();
    }
    \endcode
*/
template<class Random = MersenneTwister >
class Sampler
{
  public:
        /** Internal type of the indices.
            Currently, 64-bit indices are not supported because this
            requires extension of the random number generator classes.
        */
    typedef Int32                               IndexType;
    
    typedef ArrayVector<IndexType>              IndexArrayType;
    
        /** Type of the array view object that is returned by 
            sampledIndices() and oobIndices().
        */
    typedef ArrayVectorView <IndexType>         IndexArrayViewType;

  private:
    typedef std::map<IndexType, IndexArrayType> StrataIndicesType;
    typedef std::map<IndexType, int>            StrataSizesType;
    typedef ArrayVector<bool>                   IsUsedArrayType;
    typedef ArrayVectorView<bool>               IsUsedArrayViewType;
    
    static const int        oobInvalid = -1;

    int                     total_count_, sample_size_;
    mutable int             current_oob_count_;
    StrataIndicesType       strata_indices_;
    StrataSizesType         strata_sample_size_;
    IndexArrayType          current_sample_;
    mutable IndexArrayType  current_oob_sample_;
    IsUsedArrayType         is_used_;
    Random                  default_random_;
    Random const &          random_;
    SamplerOptions          options_;

    void initStrataCount()
    {
        // compute how many samples to take from each stratum
        // (may be unequal if sample_size_ is not a multiple of strataCount())
        int strata_sample_size = static_cast<int>(std::ceil(double(sample_size_) / strataCount()));
        int strata_total_count = strata_sample_size * strataCount();

        for(StrataIndicesType::iterator i = strata_indices_.begin(); 
             i != strata_indices_.end(); ++i)
        {
            if(strata_total_count > sample_size_)
            {
                strata_sample_size_[i->first] = strata_sample_size - 1;
                --strata_total_count;
            }
            else
            {
                strata_sample_size_[i->first] = strata_sample_size;
            }
        }
    }

  public:
    
        /** Create a sampler for \a totalCount data objects.
            
            In each invocation of <tt>sample()</tt> below, it will sample
            indices according to the options passed. If no options are given, 
            <tt>totalCount</tt> indices will be drawn with replacement.
        */
    Sampler(UInt32 totalCount, SamplerOptions const & opt = SamplerOptions(), 
            Random const * rnd = 0)
    : total_count_(totalCount),
      sample_size_(opt.sample_size == 0
                   ? static_cast<int>((std::ceil(total_count_ * opt.sample_proportion)))
                   : opt.sample_size),
      current_oob_count_(oobInvalid),
      current_sample_(sample_size_),
      current_oob_sample_(total_count_),
      is_used_(total_count_),
      default_random_(RandomSeed),
      random_(rnd ? *rnd : default_random_),
      options_(opt)
    {
        vigra_precondition(opt.sample_with_replacement || sample_size_ <= total_count_,
          "Sampler(): Cannot draw without replacement when data size is smaller than sample count.");
          
        vigra_precondition(!opt.stratified_sampling,
          "Sampler(): Stratified sampling requested, but no strata given.");
          
        // initialize a single stratum containing all data
        strata_indices_[0].resize(total_count_);
        for(int i=0; i<total_count_; ++i)
            strata_indices_[0][i] = i;

        initStrataCount();
        //this is screwing up the random forest tests.
        //sample();
    }
    
        /** Create a sampler for stratified sampling.
            
            <tt>strataBegin</tt> and <tt>strataEnd</tt> must refer to a sequence 
            which specifies for each sample the stratum it belongs to. The
            total number of data objects will be set to <tt>strataEnd - strataBegin</tt>.
            Equally many samples (subject to rounding) will be drawn from each stratum, 
            unless the option object explicitly requests unstratified sampling, 
            in which case the strata are ignored.
        */
    template <class Iterator>
    Sampler(Iterator strataBegin, Iterator strataEnd, SamplerOptions const & opt = SamplerOptions(), 
            Random const * rnd = 0)
    : total_count_(strataEnd - strataBegin),
      sample_size_(opt.sample_size == 0
                   ? static_cast<int>((std::ceil(total_count_ * opt.sample_proportion)))
                   : opt.sample_size),
      current_oob_count_(oobInvalid),
      current_sample_(sample_size_),
      current_oob_sample_(total_count_),
      is_used_(total_count_),
      default_random_(RandomSeed),
      random_(rnd ? *rnd : default_random_),
      options_(opt)
    {
        vigra_precondition(opt.sample_with_replacement || sample_size_ <= total_count_,
          "Sampler(): Cannot draw without replacement when data size is smaller than sample count.");
          
        // copy the strata indices
        if(opt.stratified_sampling)
        {
            for(int i = 0; strataBegin != strataEnd; ++i, ++strataBegin)
            {
                strata_indices_[*strataBegin].push_back(i);
            }
        }
        else
        {
            strata_indices_[0].resize(total_count_);
            for(int i=0; i<total_count_; ++i)
                strata_indices_[0][i] = i;
        }
            
        vigra_precondition(sample_size_ >= static_cast<int>(strata_indices_.size()),
            "Sampler(): Requested sample count must be at least as large as the number of strata.");

        initStrataCount();
        //this is screwing up the random forest tests.
        //sample();
    }

        /** Return the k-th index in the current sample.
         */
    IndexType operator[](int k) const
    {
        return current_sample_[k];
    }

        /** Create a new sample.
         */
    void sample();

        /** The total number of data elements.
         */
    int totalCount() const
    {
        return total_count_;
    }

        /** The number of data elements that have been sampled.
         */
    int sampleSize() const
    {
        return sample_size_;
    }

        /** Same as sampleSize().
         */
    int size() const
    {
        return sample_size_;
    }

        /** The number of strata to be used.
            Will be 1 if no strata are given. Will be ignored when
            stratifiedSampling() is false.
         */
    int strataCount() const
    {
        return strata_indices_.size();
    }

        /** Whether to use stratified sampling.
            (If this is false, strata will be ignored even if present.)
         */
    bool stratifiedSampling() const
    {
        return options_.stratified_sampling;
    }
    
        /** Whether sampling should be performed with replacement.
         */
    bool withReplacement() const
    {
        return options_.sample_with_replacement;
    }
    
        /** Return an array view containing the indices in the current sample.
         */
    IndexArrayViewType sampledIndices() const
    {
        return current_sample_;
    }
    
        /** Return an array view containing the out-of-bag indices.
            (i.e. the indices that are not in the current sample)
         */
    IndexArrayViewType oobIndices() const
    {
        if(current_oob_count_ == oobInvalid)
        {
            current_oob_count_ = 0;
            for(int i = 0; i<total_count_; ++i)
            {
                if(!is_used_[i])
                {
                    current_oob_sample_[current_oob_count_] = i;
                    ++current_oob_count_;
                }
            }
        }
        return current_oob_sample_.subarray(0, current_oob_count_);
    }
    IsUsedArrayType const & is_used() const
    {
        return is_used_;
    }
};


template<class Random>
void Sampler<Random>::sample()
{
    current_oob_count_ = oobInvalid;
    is_used_.init(false);
    
    if(options_.sample_with_replacement)
    {
        //Go thru all strata
        int j = 0;
        StrataIndicesType::iterator iter;
        for(iter = strata_indices_.begin(); iter != strata_indices_.end(); ++iter)
        {
            // do sampling with replacement in each strata and copy data.
            int stratum_size = iter->second.size();
            for(int i = 0; i < static_cast<int>(strata_sample_size_[iter->first]); ++i, ++j)
            {
                current_sample_[j] = iter->second[random_.uniformInt(stratum_size)];
                is_used_[current_sample_[j]] = true;
            }
        }
    }
    else
    {
        //Go thru all strata
        int j = 0;
        StrataIndicesType::iterator iter;
        for(iter = strata_indices_.begin(); iter != strata_indices_.end(); ++iter)
        {
            // do sampling without replacement in each strata and copy data.
            int stratum_size = iter->second.size();
            for(int i = 0; i < static_cast<int>(strata_sample_size_[iter->first]); ++i, ++j)
            {
                std::swap(iter->second[i], iter->second[i+ random_.uniformInt(stratum_size - i)]);
                current_sample_[j] = iter->second[i];
                is_used_[current_sample_[j]] = true;
            }
        }
    }
}

template<class Random =RandomTT800 >
class PoissonSampler
{
public:
    Random  randfloat;
    typedef Int32                               IndexType;
    typedef vigra::ArrayVector     <IndexType>  IndexArrayType;
    IndexArrayType        used_indices_;
    double lambda;
    int minIndex;
    int maxIndex;
    
    PoissonSampler(double lambda,IndexType minIndex,IndexType maxIndex)
    : lambda(lambda),
      minIndex(minIndex),
      maxIndex(maxIndex)
    {}

    void sample(  )
    {
        used_indices_.clear();
        IndexType i;
        for(i=minIndex;i<maxIndex;++i)
        {
            //from http://en.wikipedia.org/wiki/Poisson_distribution
            int k=0;
            double p=1;
            double L=exp(-lambda);
            do
            {
                ++k;
                p*=randfloat.uniform53();

            }while(p>L);
            --k;
            //Insert i this many time
            while(k>0)
            {
                used_indices_.push_back(i);
                --k;
            }
        }
    }

    IndexType const & operator[](int in) const
    {
        return used_indices_[in];
    }
    
    int numOfSamples() const
    {
        return used_indices_.size();
    }
};

//@}

} // namespace vigra

#endif /*VIGRA_SAMPLING_HXX*/