X-Git-Url: https://pintos-os.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=src%2Fmath%2Fcovariance.c;h=2500f90423b6b2334841b6a95ae21fe6b3d66586;hb=bb0e1c1f8b1a87a0ccd9bb4ffb22466475a47b7f;hp=b5a4166dfa40643c5a71c88f37ceb7b4c604b1d5;hpb=2b5831d68df1824881f9117bf2297957a6e49f64;p=pspp diff --git a/src/math/covariance.c b/src/math/covariance.c index b5a4166dfa..2500f90423 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; @@ -156,8 +160,8 @@ covariance_1pass_create (size_t n_vars, const struct variable **vars, cov->n_cm = (n_vars * (n_vars - 1) ) / 2; - if (cov->n_cm > 0) - cov->cm = xcalloc (sizeof *cov->cm, cov->n_cm); + + cov->cm = xcalloc (cov->n_cm, sizeof *cov->cm); cov->categoricals = NULL; return cov; @@ -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); @@ -264,7 +268,7 @@ get_val (const struct covariance *cov, int i, const struct ccase *c) return val->f; } - return categoricals_get_binary_by_subscript (cov->categoricals, i - cov->n_vars, c); + return categoricals_get_effects_code_for_case (cov->categoricals, i - cov->n_vars, c); } #if 0 @@ -296,7 +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) { @@ -342,11 +347,16 @@ 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); + cov->cm = xcalloc (cov->n_cm, sizeof *cov->cm); /* Grow the moment matrices so that they're large enough to accommodate the categorical elements */ @@ -355,8 +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) { @@ -524,7 +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; @@ -548,7 +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; @@ -593,12 +601,12 @@ 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) { if ( cov->state <= 0 ) @@ -620,29 +628,13 @@ 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; - for (i = 0 ; i < cov->dim; ++i) - { - for (j = 0 ; j < cov->dim; ++j) - { - int idx; - double *x = gsl_matrix_ptr (cov->moments[MOMENT_VARIANCE], i, j); - - idx = cm_idx (cov, i, j); - if ( idx >= 0) - { - x = &cov->cm [idx]; - } - } - } - return cm_to_gsl (cov); } -static const gsl_matrix * +static gsl_matrix * covariance_calculate_single_pass_unnormalized (struct covariance *cov) { size_t i, j; @@ -674,12 +666,12 @@ 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) { if ( cov->state <= 0 ) @@ -698,6 +690,14 @@ 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 */ @@ -715,3 +715,96 @@ covariance_destroy (struct covariance *cov) free (cov->cm); free (cov); } + +size_t +covariance_dim (const struct covariance * cov) +{ + return (cov->dim); +} + + + +/* + Routines to assist debugging. + The following are not thoroughly tested and in certain respects + unreliable. They should only be + used for aids to development. Not as user accessible code. +*/ + +#include "libpspp/str.h" +#include "output/tab.h" +#include "data/format.h" + + +/* Create a table which can be populated with the encodings for + the covariance matrix COV */ +struct tab_table * +covariance_dump_enc_header (const struct covariance *cov, int length) +{ + struct tab_table *t = tab_create (cov->dim, length); + int n; + int i; + + tab_title (t, "Covariance Encoding"); + + tab_box (t, + TAL_2, TAL_2, 0, 0, + 0, 0, tab_nc (t) - 1, tab_nr (t) - 1); + + tab_hline (t, TAL_2, 0, tab_nc (t) - 1, 1); + + + for (i = 0 ; i < cov->n_vars; ++i) + { + tab_text (t, i, 0, TAT_TITLE, var_get_name (cov->vars[i])); + tab_vline (t, TAL_1, i + 1, 0, tab_nr (t) - 1); + } + + n = 0; + while (i < cov->dim) + { + struct string str; + int idx = i - cov->n_vars; + const struct interaction *iact = + categoricals_get_interaction_by_subscript (cov->categoricals, idx); + int df; + + ds_init_empty (&str); + interaction_to_string (iact, &str); + + df = categoricals_df (cov->categoricals, n); + + tab_joint_text (t, + i, 0, + i + df - 1, 0, + TAT_TITLE, ds_cstr (&str)); + + if (i + df < tab_nr (t) - 1) + tab_vline (t, TAL_1, i + df, 0, tab_nr (t) - 1); + + i += df; + n++; + ds_destroy (&str); + } + + return t; +} + + +/* + Append table T, which should have been returned by covariance_dump_enc_header + with an entry corresponding to case C for the covariance matrix COV + */ +void +covariance_dump_enc (const struct covariance *cov, const struct ccase *c, + struct tab_table *t) +{ + static int row = 0; + int i; + ++row; + for (i = 0 ; i < cov->dim; ++i) + { + double v = get_val (cov, i, c); + tab_double (t, i, row, 0, v, i < cov->n_vars ? NULL : &F_8_0); + } +}