X-Git-Url: https://pintos-os.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=src%2Fregression.q;h=660c15d82338647bef0bab3a9a33fadc5357f5e6;hb=012b055f5f7b4ddd2a608e64c53eaae101b5634e;hp=ee7c0075442288cc32afb9e2aa9f8ff48982cd64;hpb=c38b1e667a1fdb12df302ce54872dee88c04e65e;p=pspp

diff --git a/src/regression.q b/src/regression.q
index ee7c007544..660c15d823 100644
--- a/src/regression.q
+++ b/src/regression.q
@@ -26,7 +26,9 @@
 #include "case.h"
 #include "casefile.h"
 #include "cat.h"
+#include "cat-routines.h"
 #include "command.h"
+#include "design-matrix.h"
 #include "dictionary.h"
 #include "error.h"
 #include "file-handle.h"
@@ -34,12 +36,14 @@
 #include "lexer.h"
 #include <linreg/pspp_linreg.h>
 #include "missing-values.h"
+#include "reg_export_comments.h"
 #include "tab.h"
 #include "var.h"
 #include "vfm.h"
 
-/* (headers) */
+#define REG_LARGE_DATA 1000
 
+/* (headers) */
 
 /* (specification)
    "REGRESSION" (regression_):
@@ -61,6 +65,7 @@
    f,
    defaults,
    all;
+   export=custom;
    ^dependent=varlist;
    ^method=enter.
 */
@@ -73,6 +78,12 @@ static struct cmd_regression cmd;
  */
 size_t *indep_vars;
 
+/*
+  File where the model will be saved if the EXPORT subcommand
+  is given. 
+ */
+struct file_handle *model_file;
+
 /*
   Return value for the procedure.
  */
@@ -461,6 +472,69 @@ subcommand_statistics (int *keywords, pspp_linreg_cache * c)
   statistics_keyword_output (reg_stats_tol, keywords[tol], c);
   statistics_keyword_output (reg_stats_selection, keywords[selection], c);
 }
+static void
+subcommand_export (int export, pspp_linreg_cache *c)
+{
+  FILE *fp;
+  size_t i;
+  struct pspp_linreg_coeff coeff;
+
+  if (export)
+    {
+      assert (c != NULL);
+      assert (model_file != NULL);
+      assert (fp != NULL);
+      fp = fopen (handle_get_filename (model_file), "w");
+      fprintf (fp, "#include <string.h>\n\n");
+      fprintf (fp, "%s", reg_mean_cmt);
+      fprintf (fp, "double\npspp_reg_estimate (const double *var_vals, const char *var_names[])\n{\n\tchar *model_depvars[%d] = {", c->n_indeps);
+      for (i = 1; i < c->n_indeps; i++)
+	{
+	  coeff = c->coeff[i];
+	  fprintf (fp, "\"%s\",\n\t\t", coeff.v->name);
+	}
+      coeff = c->coeff[i];
+      fprintf (fp, "\"%s\"};\n\t", coeff.v->name);
+      fprintf (fp, "double model_coeffs[%d] = {", c->n_indeps);
+      for (i = 1; i < c->n_indeps; i++)
+	{
+	  coeff = c->coeff[i];
+	  fprintf (fp, "%.15e,\n\t\t", coeff.estimate);
+	}
+      coeff = c->coeff[i];
+      fprintf (fp, "%.15e};\n\t", coeff.estimate);
+      coeff = c->coeff[0];
+      fprintf (fp, "double estimate = %.15e;\n\t", coeff.estimate);
+      fprintf (fp, "int i;\n\tint j;\n\n\t");
+      fprintf (fp, "for (i = 0; i < %d; i++)\n\t", c->n_indeps);
+      fprintf (fp, "{\n\t\tfor (j = 0; j < %d; j++)\n\t\t", c->n_indeps);
+      fprintf (fp, "{\n\t\t\tif (strcmp (var_names[i], model_depvars[j]) == 0)\n");
+      fprintf (fp, "\t\t\t{\n\t\t\t\testimate += var_vals[i] * model_coeffs[j];\n");
+      fprintf (fp, "\t\t\t}\n\t\t}\n\t}\n\treturn estimate;\n}\n");
+      fclose (fp);
+    }
+}
+static int
+regression_custom_export (struct cmd_regression *cmd)
+{
+  /* 0 on failure, 1 on success, 2 on failure that should result in syntax error */
+  if (!lex_force_match ('('))
+    return 0;
+  
+  if (lex_match ('*'))
+    model_file = NULL;
+  else 
+    {
+      model_file = fh_parse ();
+      if (model_file == NULL)
+        return 0; 
+    }
+  
+  if (!lex_force_match (')'))
+    return 0;
+
+  return 1;
+}
 
 int
 cmd_regression (void)
@@ -497,7 +571,6 @@ static void
 run_regression (const struct casefile *cf, void *cmd_ UNUSED)
 {
   size_t i;
-  size_t k;
   size_t n_data = 0;
   size_t row;
   size_t case_num;
@@ -512,6 +585,7 @@ run_regression (const struct casefile *cf, void *cmd_ UNUSED)
   struct casereader *r2;
   struct ccase c;
   struct variable *v;
+  struct variable **indep_vars;
   struct design_matrix *X;
   gsl_vector *Y;
   pspp_linreg_cache *lcache;
@@ -534,29 +608,35 @@ run_regression (const struct casefile *cf, void *cmd_ UNUSED)
      Read from the active file. The first pass encodes categorical
      variables and drops cases with missing values.
    */
+  j = 0;
   for (i = 0; i < cmd.n_variables; i++)
     {
-      v = cmd.v_variables[i];
-      if (v->type == ALPHA)
+      if (!is_depvar (i))
 	{
-	  /* Make a place to hold the binary vectors 
-	     corresponding to this variable's values. */
-	  cat_stored_values_create (v);
-	}
-      for (r = casefile_get_reader (cf);
-	   casereader_read (r, &c); case_destroy (&c))
-	{
-	  row = casereader_cnum (r) - 1;
-
-	  val = case_data (&c, v->fv);
-	  cat_value_update (v, val);
-	  if (mv_is_value_missing (&v->miss, val))
+	  v = cmd.v_variables[i];
+	  indep_vars[j] = v;
+	  j++;
+	  if (v->type == ALPHA)
 	    {
-	      if (!is_missing_case[row])
+	      /* Make a place to hold the binary vectors 
+		 corresponding to this variable's values. */
+	      cat_stored_values_create (v);
+	    }
+	  for (r = casefile_get_reader (cf);
+	       casereader_read (r, &c); case_destroy (&c))
+	    {
+	      row = casereader_cnum (r) - 1;
+	      
+	      val = case_data (&c, v->fv);
+	      cat_value_update (v, val);
+	      if (mv_is_value_missing (&v->miss, val))
 		{
-		  /* Now it is missing. */
-		  n_data--;
-		  is_missing_case[row] = 1;
+		  if (!is_missing_case[row])
+		    {
+		      /* Now it is missing. */
+		      n_data--;
+		      is_missing_case[row] = 1;
+		    }
 		}
 	    }
 	}
@@ -564,7 +644,7 @@ run_regression (const struct casefile *cf, void *cmd_ UNUSED)
 
   Y = gsl_vector_alloc (n_data);
   X =
-    design_matrix_create (n_indep, (const struct variable **) cmd.v_variables,
+    design_matrix_create (n_indep, (const struct variable **) indep_vars,
 			  n_data);
   lcache = pspp_linreg_cache_alloc (X->m->size1, X->m->size2);
   lcache->indep_means = gsl_vector_alloc (X->m->size2);
@@ -578,7 +658,6 @@ run_regression (const struct casefile *cf, void *cmd_ UNUSED)
        case_destroy (&c))
     /* Iterate over the cases. */
     {
-      k = 0;
       case_num = casereader_cnum (r2) - 1;
       if (!is_missing_case[case_num])
 	{
@@ -604,7 +683,7 @@ run_regression (const struct casefile *cf, void *cmd_ UNUSED)
 		      pspp_reg_rc = CMD_FAILURE;
 		      return;
 		    }
-		  lcache->depvar = (const struct var *) v;
+		  lcache->depvar = (const struct variable *) v;
 		  gsl_vector_set (Y, row, val->f);
 		}
 	      else
@@ -618,8 +697,6 @@ run_regression (const struct casefile *cf, void *cmd_ UNUSED)
 		      design_matrix_set_numeric (X, row, v, val);
 		    }
 
-		  indep_vars[k] = i;
-		  k++;
 		  lopts.get_indep_mean_std[i] = 1;
 		}
 	    }
@@ -639,11 +716,19 @@ run_regression (const struct casefile *cf, void *cmd_ UNUSED)
 	(const struct variable *) design_matrix_col_to_var (X, i);
       assert (lcache->coeff[j].v != NULL);
     }
+  /*
+    For large data sets, use QR decomposition.
+   */
+  if (n_data > sqrt (n_indep) && n_data > REG_LARGE_DATA)
+    {
+      lcache->method = PSPP_LINREG_SVD;
+    }
   /* 
      Find the least-squares estimates and other statistics.
    */
   pspp_linreg ((const gsl_vector *) Y, X->m, &lopts, lcache);
   subcommand_statistics (cmd.a_statistics, lcache);
+  subcommand_export (cmd.sbc_export, lcache);
   gsl_vector_free (Y);
   design_matrix_destroy (X);
   pspp_linreg_cache_free (lcache);