1 /* PSPP - a program for statistical analysis.
2 Copyright (C) 2017 Free Software Foundation, Inc.
4 This program is free software: you can redistribute it and/or modify
5 it under the terms of the GNU General Public License as published by
6 the Free Software Foundation, either version 3 of the License, or
7 (at your option) any later version.
9 This program is distributed in the hope that it will be useful,
10 but WITHOUT ANY WARRANTY; without even the implied warranty of
11 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
12 GNU General Public License for more details.
14 You should have received a copy of the GNU General Public License
15 along with this program. If not, see <http://www.gnu.org/licenses/>. */
19 #include "matrix-reader.h"
23 #include <libpspp/message.h>
24 #include <data/casegrouper.h>
25 #include <data/casereader.h>
26 #include <data/dictionary.h>
27 #include <data/variable.h>
30 #define _(msgid) gettext (msgid)
31 #define N_(msgid) msgid
35 This module interprets a "data matrix", typically generated by the command
36 MATRIX DATA. The dictionary of such a matrix takes the form:
38 s_0, s_1, ... s_m, ROWTYPE_, VARNAME_, v_0, v_1, .... v_n
40 where s_0, s_1 ... s_m are the variables defining the splits, and
41 v_0, v_1 ... v_n are the continuous variables.
45 The ROWTYPE_ variable is of type A8.
46 The VARNAME_ variable is a string type whose width is not predetermined.
47 The variables s_x are of type F4.0 (although this reader accepts any type),
48 and v_x are of any numeric type.
50 The values of the ROWTYPE_ variable are in the set {MEAN, STDDEV, N, CORR, COV}
51 and determine the purpose of that case.
52 The values of the VARNAME_ variable must correspond to the names of the varibles
53 in {v_0, v_1 ... v_n} and indicate the rows of the correlation or covariance
58 A typical example is as follows:
60 s_0 ROWTYPE_ VARNAME_ v_0 v_1 v_2
62 0 MEAN 5.0000 4.0000 3.0000
63 0 STDDEV 1.0000 2.0000 3.0000
64 0 N 9.0000 9.0000 9.0000
65 0 CORR V1 1.0000 .6000 .7000
66 0 CORR V2 .6000 1.0000 .8000
67 0 CORR V3 .7000 .8000 1.0000
68 1 MEAN 9.0000 8.0000 7.0000
69 1 STDDEV 5.0000 6.0000 7.0000
70 1 N 9.0000 9.0000 9.0000
71 1 CORR V1 1.0000 .4000 .3000
72 1 CORR V2 .4000 1.0000 .2000
73 1 CORR V3 .3000 .2000 1.0000
79 const struct dictionary *dict;
80 const struct variable *varname;
81 const struct variable *rowtype;
82 struct casegrouper *grouper;
84 gsl_matrix *n_vectors;
85 gsl_matrix *mean_vectors;
86 gsl_matrix *var_vectors;
88 gsl_matrix *correlation;
89 gsl_matrix *covariance;
92 struct matrix_reader *
93 create_matrix_reader_from_case_reader (const struct dictionary *dict, struct casereader *in_reader,
94 const struct variable ***vars, size_t *n_vars)
96 struct matrix_reader *mr = xzalloc (sizeof *mr);
99 mr->varname = dict_lookup_var (dict, "varname_");
100 if (mr->varname == NULL)
102 msg (ME, _("Matrix dataset lacks a variable called %s."), "VARNAME_");
107 mr->rowtype = dict_lookup_var (dict, "rowtype_");
108 if (mr->rowtype == NULL)
110 msg (ME, _("Matrix dataset lacks a variable called %s."), "ROWTYPE_");
116 const struct variable **dvars = NULL;
117 dict_get_vars (dict, &dvars, &dvarcnt, DC_SCRATCH);
120 *n_vars = dvarcnt - var_get_dict_index (mr->varname) - 1;
125 *vars = xcalloc (sizeof (struct variable **), *n_vars);
127 for (i = 0; i < *n_vars; ++i)
129 (*vars)[i] = dvars[i + var_get_dict_index (mr->varname) + 1];
133 /* All the variables before ROWTYPE_ (if any) are split variables */
134 mr->grouper = casegrouper_create_vars (in_reader, dvars, var_get_dict_index (mr->rowtype));
142 destroy_matrix_reader (struct matrix_reader *mr)
146 bool ret = casegrouper_destroy (mr->grouper);
153 next_matrix_from_reader (struct matrix_material *mm,
154 struct matrix_reader *mr,
155 const struct variable **vars, int n_vars)
157 struct casereader *group;
159 gsl_matrix_free (mr->n_vectors);
160 gsl_matrix_free (mr->mean_vectors);
161 gsl_matrix_free (mr->var_vectors);
162 gsl_matrix_free (mr->correlation);
163 gsl_matrix_free (mr->covariance);
165 if (!casegrouper_get_next_group (mr->grouper, &group))
168 mr->n_vectors = gsl_matrix_alloc (n_vars, n_vars);
169 mr->mean_vectors = gsl_matrix_alloc (n_vars, n_vars);
170 mr->var_vectors = gsl_matrix_alloc (n_vars, n_vars);
172 mm->n = mr->n_vectors;
173 mm->mean_matrix = mr->mean_vectors;
174 mm->var_matrix = mr->var_vectors;
176 mr->correlation = NULL;
177 mr->covariance = NULL;
181 for ( ; (c = casereader_read (group) ); case_unref (c))
183 const union value *uv = case_data (c, mr->rowtype);
185 for (col = 0; col < n_vars; ++col)
187 const struct variable *cv
188 = vars ? vars[col] : dict_get_var (mr->dict, var_get_dict_index (mr->varname) + 1 + col);
189 double x = case_data (c, cv)->f;
190 if (0 == strncasecmp ((char *)value_str (uv, 8), "N ", 8))
191 for (row = 0; row < n_vars; ++row)
192 gsl_matrix_set (mr->n_vectors, row, col, x);
193 else if (0 == strncasecmp ((char *) value_str (uv, 8), "MEAN ", 8))
194 for (row = 0; row < n_vars; ++row)
195 gsl_matrix_set (mr->mean_vectors, row, col, x);
196 else if (0 == strncasecmp ((char *) value_str (uv, 8), "STDDEV ", 8))
197 for (row = 0; row < n_vars; ++row)
198 gsl_matrix_set (mr->var_vectors, row, col, x * x);
200 if (0 == strncasecmp ((char *) value_str (uv, 8), "CORR ", 8))
202 if (mr->correlation == NULL)
203 mr->correlation = gsl_matrix_alloc (n_vars, n_vars);
204 for (col = 0; col < n_vars; ++col)
206 const struct variable *cv
207 = vars ? vars[col] : dict_get_var (mr->dict, var_get_dict_index (mr->varname) + 1 + col);
208 double x = case_data (c, cv)->f;
209 gsl_matrix_set (mr->correlation, crow, col, x);
213 else if (0 == strncasecmp ((char *) value_str (uv, 8), "COV ", 8))
215 if (mr->covariance == NULL)
216 mr->covariance = gsl_matrix_alloc (n_vars, n_vars);
217 for (col = 0; col < n_vars; ++col)
219 const struct variable *cv
220 = vars ? vars[col] : dict_get_var (mr->dict, var_get_dict_index (mr->varname) + 1 + col);
221 double x = case_data (c, cv)->f;
222 gsl_matrix_set (mr->covariance, crow, col, x);
228 casereader_destroy (group);
230 mm->cov = mr->covariance;
231 mm->corr = mr->correlation;