Line data    Source code

       1             : /**
       2             :  * @file gensvm_checks.c
       3             :  * @author G.J.J. van den Burg
       4             :  * @date 2016-12-07
       5             :  * @brief Sanity checks used to ensure inputs are as expected
       6             :  *
       7             :  * @copyright
       8             :  Copyright 2016, G.J.J. van den Burg.
       9             : 
      10             :  This file is part of GenSVM.
      11             : 
      12             :  GenSVM is free software: you can redistribute it and/or modify
      13             :  it under the terms of the GNU General Public License as published by
      14             :  the Free Software Foundation, either version 3 of the License, or
      15             :  (at your option) any later version.
      16             : 
      17             :  GenSVM is distributed in the hope that it will be useful,
      18             :  but WITHOUT ANY WARRANTY; without even the implied warranty of
      19             :  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
      20             :  GNU General Public License for more details.
      21             : 
      22             :  You should have received a copy of the GNU General Public License
      23             :  along with GenSVM. If not, see <http://www.gnu.org/licenses/>.
      24             : 
      25             :  */
      26             : 
      27             : #include "gensvm_checks.h"
      28             : 
      29             : /**
      30             :  * @brief Check if the labels are contiguous on [1 .. K]
      31             :  *
      32             :  * @details
      33             :  * The GenSVM library currently requires that the labels that are supplied in 
      34             :  * a dataset are contigous on the interval [1 .. K] and have no gaps. This is 
      35             :  * required because the dimensionality of the problem is directly related to 
      36             :  * the maximum class label K. This function checks if the labels are indeed in 
      37             :  * the desired range.
      38             :  *
      39             :  * @param[in]   data    a GenData struct with the current data
      40             :  *
      41             :  * @return              whether the labels are contiguous or not
      42             :  */
      43           4 : bool gensvm_check_outcome_contiguous(struct GenData *data)
      44             : {
      45           4 :         bool in_uniq, is_contiguous = true;
      46           4 :         long i, j, K = 1;
      47           4 :         long max_y = -1,
      48           4 :              min_y = LONG_MAX;
      49           4 :         long *uniq_y = Calloc(long, K);
      50           4 :         uniq_y[0] = data->y[0];
      51             : 
      52          40 :         for (i=1; i<data->n; i++) {
      53          36 :                 in_uniq = false;
      54          87 :                 for (j=0; j<K; j++) {
      55          77 :                         if (uniq_y[j] == data->y[i]) {
      56          26 :                                 in_uniq = true;
      57          26 :                                 break;
      58             :                         }
      59             :                 }
      60             : 
      61          36 :                 if (!in_uniq) {
      62          10 :                         uniq_y = Realloc(uniq_y, long, K+1);
      63          10 :                         uniq_y[K++] = data->y[i];
      64             :                 }
      65             : 
      66          36 :                 max_y = maximum(max_y, data->y[i]);
      67          36 :                 min_y = minimum(min_y, data->y[i]);
      68             :         }
      69             : 
      70           4 :         if (min_y < 1 || max_y > K) {
      71           3 :                 is_contiguous = false;
      72             :         }
      73             : 
      74           4 :         free(uniq_y);
      75             : 
      76           4 :         return is_contiguous;
      77             : }