X-Git-Url: https://pintos-os.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=src%2Fmath%2Fcovariance.c;h=f64e89d1eb2ffbee673c2179e6d2e48b78d066f3;hb=32ee0e0402d6d56674f53a47d879ec5c07dabe09;hp=d03f6ad07a3039955047a687d7c9115cb99178c1;hpb=930138c8279976294019300966e1fb842173f762;p=pspp diff --git a/src/math/covariance.c b/src/math/covariance.c index d03f6ad07a..f64e89d1eb 100644 --- a/src/math/covariance.c +++ b/src/math/covariance.c @@ -1,5 +1,5 @@ /* PSPP - a program for statistical analysis. - Copyright (C) 2009 Free Software Foundation, Inc. + Copyright (C) 2009, 2010, 2011 Free Software Foundation, Inc. This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by @@ -16,15 +16,19 @@ #include -#include -#include "covariance.h" -#include -#include "moments.h" +#include "math/covariance.h" + #include -#include -#include -#include -#include "categoricals.h" + +#include "data/case.h" +#include "data/variable.h" +#include "libpspp/assertion.h" +#include "libpspp/misc.h" +#include "math/categoricals.h" +#include "math/interaction.h" +#include "math/moments.h" + +#include "gl/xalloc.h" #define n_MOMENTS (MOMENT_VARIANCE + 1) @@ -68,7 +72,7 @@ struct covariance { /* The variables for which the covariance matrix is to be calculated. */ size_t n_vars; - const struct variable **vars; + const struct variable *const *vars; /* Categorical variables. */ struct categoricals *categoricals; @@ -131,7 +135,7 @@ covariance_moments (const struct covariance *cov, int m) /* Create a covariance struct. */ struct covariance * -covariance_1pass_create (size_t n_vars, const struct variable **vars, +covariance_1pass_create (size_t n_vars, const struct variable *const *vars, const struct variable *weight, enum mv_class exclude) { size_t i; @@ -170,8 +174,8 @@ covariance_1pass_create (size_t n_vars, const struct variable **vars, until then. */ struct covariance * -covariance_2pass_create (size_t n_vars, const struct variable **vars, - size_t n_catvars, const struct variable **catvars, +covariance_2pass_create (size_t n_vars, const struct variable *const *vars, + struct categoricals *cats, const struct variable *wv, enum mv_class exclude) { size_t i; @@ -197,7 +201,7 @@ covariance_2pass_create (size_t n_vars, const struct variable **vars, cov->n_cm = -1; cov->cm = NULL; - cov->categoricals = categoricals_create (catvars, n_catvars, wv, exclude); + cov->categoricals = cats; return cov; } @@ -244,7 +248,7 @@ is_missing (const struct covariance *cov, int i, const struct ccase *c) { const struct variable *var = i < cov->n_vars ? cov->vars[i] : - categoricals_get_variable_by_subscript (cov->categoricals, i - cov->n_vars); + categoricals_get_interaction_by_subscript (cov->categoricals, i - cov->n_vars)->vars[0]; const union value *val = case_data (c, var); @@ -267,6 +271,7 @@ get_val (const struct covariance *cov, int i, const struct ccase *c) return categoricals_get_binary_by_subscript (cov->categoricals, i - cov->n_vars, c); } +#if 0 void dump_matrix (const gsl_matrix *m) { @@ -279,6 +284,7 @@ dump_matrix (const gsl_matrix *m) printf ("\n"); } } +#endif /* Call this function for every case in the data set */ void @@ -294,7 +300,8 @@ covariance_accumulate_pass1 (struct covariance *cov, const struct ccase *c) cov->state = 1; } - categoricals_update (cov->categoricals, c); + if (cov->categoricals) + categoricals_update (cov->categoricals, c); for (i = 0 ; i < cov->dim; ++i) { @@ -340,8 +347,13 @@ covariance_accumulate_pass2 (struct covariance *cov, const struct ccase *c) assert (cov->state == 1); cov->state = 2; - cov->dim = cov->n_vars + - categoricals_total (cov->categoricals) - categoricals_get_n_variables (cov->categoricals); + if (cov->categoricals) + categoricals_done (cov->categoricals); + + cov->dim = cov->n_vars; + + if (cov->categoricals) + cov->dim += categoricals_df_total (cov->categoricals); cov->n_cm = (cov->dim * (cov->dim - 1) ) / 2; cov->cm = xcalloc (sizeof *cov->cm, cov->n_cm); @@ -353,8 +365,6 @@ covariance_accumulate_pass2 (struct covariance *cov, const struct ccase *c) cov->moments[i] = resize_matrix (cov->moments[i], cov->dim); } - categoricals_done (cov->categoricals); - /* Populate the moments matrices with the categorical value elements */ for (i = cov->n_vars; i < cov->dim; ++i) { @@ -522,7 +532,7 @@ cm_to_gsl (struct covariance *cov) } -static const gsl_matrix * +static gsl_matrix * covariance_calculate_double_pass (struct covariance *cov) { size_t i, j; @@ -546,7 +556,7 @@ covariance_calculate_double_pass (struct covariance *cov) return cm_to_gsl (cov); } -static const gsl_matrix * +static gsl_matrix * covariance_calculate_single_pass (struct covariance *cov) { size_t i, j; @@ -591,15 +601,16 @@ covariance_calculate_single_pass (struct covariance *cov) } -/* - Return a pointer to gsl_matrix containing the pairwise covariances. - The matrix remains owned by the COV object, and must not be freed. - Call this function only after all data have been accumulated. -*/ -const gsl_matrix * +/* Return a pointer to gsl_matrix containing the pairwise covariances. The + caller owns the returned matrix and must free it when it is no longer + needed. + + Call this function only after all data have been accumulated. */ +gsl_matrix * covariance_calculate (struct covariance *cov) { - assert ( cov->state > 0 ); + if ( cov->state <= 0 ) + return NULL; switch (cov->passes) { @@ -617,7 +628,7 @@ covariance_calculate (struct covariance *cov) /* Covariance computed without dividing by the sample size. */ -static const gsl_matrix * +static gsl_matrix * covariance_calculate_double_pass_unnormalized (struct covariance *cov) { size_t i, j; @@ -639,11 +650,10 @@ covariance_calculate_double_pass_unnormalized (struct covariance *cov) return cm_to_gsl (cov); } -static const gsl_matrix * +static gsl_matrix * covariance_calculate_single_pass_unnormalized (struct covariance *cov) { size_t i, j; - size_t m; for (i = 0 ; i < cov->dim; ++i) { @@ -672,15 +682,16 @@ covariance_calculate_single_pass_unnormalized (struct covariance *cov) } -/* - Return a pointer to gsl_matrix containing the pairwise covariances. - The matrix remains owned by the COV object, and must not be freed. - Call this function only after all data have been accumulated. -*/ -const gsl_matrix * +/* Return a pointer to gsl_matrix containing the pairwise covariances. The + caller owns the returned matrix and must free it when it is no longer + needed. + + Call this function only after all data have been accumulated. */ +gsl_matrix * covariance_calculate_unnormalized (struct covariance *cov) { - assert ( cov->state > 0 ); + if ( cov->state <= 0 ) + return NULL; switch (cov->passes) { @@ -695,6 +706,14 @@ covariance_calculate_unnormalized (struct covariance *cov) } } +/* Function to access the categoricals used by COV + The return value is owned by the COV +*/ +const struct categoricals * +covariance_get_categoricals (const struct covariance *cov) +{ + return cov->categoricals; +} /* Destroy the COV object */ @@ -712,3 +731,11 @@ covariance_destroy (struct covariance *cov) free (cov->cm); free (cov); } + +size_t +covariance_dim (const struct covariance * cov) +{ + return (cov->dim); +} + +