post_process.c

Go to the documentation of this file.

/*************************************************************************
 * Copyright (c) 2011 AT&T Intellectual Property
 * All rights reserved. This program and the accompanying materials
 * are made available under the terms of the Eclipse Public License v2.0
 * which accompanies this distribution, and is available at
 * https://www.eclipse.org/org/documents/epl-2.0/EPL-2.0.html
 *
 * Contributors: Details at https://graphviz.org
 *************************************************************************/
 
#include "config.h"
 
#include <assert.h>
#include <common/globals.h>
#include <common/types.h>
#include <limits.h>
#include <math.h>
#include <stdbool.h>
#include <stdlib.h>
#include <string.h>
#include <time.h>
 
#include <neatogen/call_tri.h>
#include <neatogen/overlap.h>
#include <sfdpgen/post_process.h>
#include <sfdpgen/sfdp.h>
#include <sfdpgen/sparse_solve.h>
#include <sfdpgen/spring_electrical.h>
#include <util/alloc.h>
#include <util/exit.h>
#include <util/gv_math.h>
#include <util/unused.h>
 
#define node_degree(i) (ia[(i) + 1] - ia[(i)])
 
static SparseMatrix ideal_distance_matrix(SparseMatrix A, int dim, double *x) {
  /* find the ideal distance between edges, either 1, or |N[i] \Union N[j]| -
   * |N[i] \Intersection N[j]|
   */
  double sum;
 
  assert(SparseMatrix_is_symmetric(A, false));
 
  SparseMatrix D = SparseMatrix_copy(A);
  const int *const ia = D->ia;
  const int *const ja = D->ja;
  if (D->type != MATRIX_TYPE_REAL) {
    free(D->a);
    D->type = MATRIX_TYPE_REAL;
    D->a = gv_calloc(D->nz, sizeof(double));
  }
  double *const d = D->a;
 
  int *mask = gv_calloc(D->m, sizeof(int));
  for (int i = 0; i < D->m; i++)
    mask[i] = -1;
 
  for (int i = 0; i < D->m; i++) {
    const double di = node_degree(i);
    mask[i] = i;
    for (int j = ia[i]; j < ia[i + 1]; j++) {
      if (i == ja[j])
        continue;
      mask[ja[j]] = i;
    }
    for (int j = ia[i]; j < ia[i + 1]; j++) {
      const int k = ja[j];
      if (i == k)
        continue;
      double len = di + node_degree(k);
      for (int l = ia[k]; l < ia[k + 1]; l++) {
        if (mask[ja[l]] == i)
          len--;
      }
      d[j] = len;
      assert(len > 0);
    }
  }
 
  sum = 0;
  double sumd = 0;
  int nz = 0;
  for (int i = 0; i < D->m; i++) {
    for (int j = ia[i]; j < ia[i + 1]; j++) {
      if (i == ja[j])
        continue;
      nz++;
      sum += distance(x, dim, i, ja[j]);
      sumd += d[j];
    }
  }
  sum /= nz;
  sumd /= nz;
  sum = sum / sumd;
 
  for (int i = 0; i < D->m; i++) {
    for (int j = ia[i]; j < ia[i + 1]; j++) {
      if (i == ja[j])
        continue;
      d[j] = sum * d[j];
    }
  }
 
  free(mask);
  return D;
}
 
StressMajorizationSmoother
StressMajorizationSmoother2_new(SparseMatrix A, int dim, double lambda0,
                                double *x, int ideal_dist_scheme) {
  /* use up to dist 2 neighbor. This is used in overcoming pherical effect with
     ideal distance of 2-neighbors equal graph distance etc.
   */
  int j, k, l, m = A->m, *ia = A->ia, *ja = A->ja, *iw, *jw, *id, *jd;
  double *d, *w, *lambda;
  double diag_d, diag_w, dist, s = 0, stop = 0, sbot = 0;
  SparseMatrix ID;
 
  assert(SparseMatrix_is_symmetric(A, false));
 
  ID = ideal_distance_matrix(A, dim, x);
 
  StressMajorizationSmoother sm =
      gv_alloc(sizeof(struct StressMajorizationSmoother_struct));
  sm->scaling = 1.;
  sm->data = NULL;
  sm->scheme = SM_SCHEME_NORMAL;
  sm->tol_cg = 0.01;
  sm->maxit_cg = floor(sqrt(A->m));
 
  lambda = sm->lambda = gv_calloc(m, sizeof(double));
  for (int i = 0; i < m; i++)
    sm->lambda[i] = lambda0;
  int *mask = gv_calloc(m, sizeof(int));
 
  double *avg_dist = gv_calloc(m, sizeof(double));
 
  for (int i = 0; i < m; i++) {
    avg_dist[i] = 0;
    int nz = 0;
    for (j = ia[i]; j < ia[i + 1]; j++) {
      if (i == ja[j])
        continue;
      avg_dist[i] += distance(x, dim, i, ja[j]);
      nz++;
    }
    assert(nz > 0);
    avg_dist[i] /= nz;
  }
 
  for (int i = 0; i < m; i++)
    mask[i] = -1;
 
  size_t nz = 0;
  for (int i = 0; i < m; i++) {
    mask[i] = i;
    for (j = ia[i]; j < ia[i + 1]; j++) {
      k = ja[j];
      if (mask[k] != i) {
        mask[k] = i;
        nz++;
      }
    }
    for (j = ia[i]; j < ia[i + 1]; j++) {
      k = ja[j];
      for (l = ia[k]; l < ia[k + 1]; l++) {
        if (mask[ja[l]] != i) {
          mask[ja[l]] = i;
          nz++;
        }
      }
    }
  }
 
  sm->Lw = SparseMatrix_new(m, m, nz + (size_t)m, MATRIX_TYPE_REAL, FORMAT_CSR);
  assert(sm->Lw != NULL);
  sm->Lwd =
      SparseMatrix_new(m, m, nz + (size_t)m, MATRIX_TYPE_REAL, FORMAT_CSR);
  assert(sm->Lwd != NULL);
 
  iw = sm->Lw->ia;
  jw = sm->Lw->ja;
 
  w = sm->Lw->a;
  d = sm->Lwd->a;
 
  id = sm->Lwd->ia;
  jd = sm->Lwd->ja;
  iw[0] = id[0] = 0;
 
  nz = 0;
  for (int i = 0; i < m; i++) {
    mask[i] = i + m;
    diag_d = diag_w = 0;
    for (j = ia[i]; j < ia[i + 1]; j++) {
      k = ja[j];
      if (mask[k] != i + m) {
        mask[k] = i + m;
 
        jw[nz] = k;
        if (ideal_dist_scheme == IDEAL_GRAPH_DIST) {
          dist = 1;
        } else if (ideal_dist_scheme == IDEAL_AVG_DIST) {
          dist = (avg_dist[i] + avg_dist[k]) * 0.5;
        } else if (ideal_dist_scheme == IDEAL_POWER_DIST) {
          dist = pow(distance_cropped(x, dim, i, k), .4);
        } else {
          fprintf(stderr, "ideal_dist_scheme value wrong");
          assert(0);
          graphviz_exit(1);
        }
 
        /*
          w[nz] = -1./(ia[i+1]-ia[i]+ia[ja[j]+1]-ia[ja[j]]);
          w[nz] = -2./(avg_dist[i]+avg_dist[k]);*/
        /* w[nz] = -1.;*/ /* use unit weight for now, later can try
                             1/(deg(i)+deg(k)) */
        w[nz] = -1 / (dist * dist);
 
        diag_w += w[nz];
 
        jd[nz] = k;
        /*
          d[nz] = w[nz]*distance(x,dim,i,k);
          d[nz] = w[nz]*dd[j];
          d[nz] = w[nz]*(avg_dist[i] + avg_dist[k])*0.5;
        */
        d[nz] = w[nz] * dist;
        stop += d[nz] * distance(x, dim, i, k);
        sbot += d[nz] * dist;
        diag_d += d[nz];
 
        nz++;
      }
    }
 
    /* distance 2 neighbors */
    for (j = ia[i]; j < ia[i + 1]; j++) {
      k = ja[j];
      for (l = ia[k]; l < ia[k + 1]; l++) {
        if (mask[ja[l]] != i + m) {
          mask[ja[l]] = i + m;
 
          if (ideal_dist_scheme == IDEAL_GRAPH_DIST) {
            dist = 2;
          } else if (ideal_dist_scheme == IDEAL_AVG_DIST) {
            dist = (avg_dist[i] + 2 * avg_dist[k] + avg_dist[ja[l]]) * 0.5;
          } else if (ideal_dist_scheme == IDEAL_POWER_DIST) {
            dist = pow(distance_cropped(x, dim, i, ja[l]), .4);
          } else {
            fprintf(stderr, "ideal_dist_scheme value wrong");
            assert(0);
            graphviz_exit(1);
          }
 
          jw[nz] = ja[l];
          /*
            w[nz] = -1/(ia[i+1]-ia[i]+ia[ja[l]+1]-ia[ja[l]]);
            w[nz] = -2/(avg_dist[i] + 2*avg_dist[k] + avg_dist[ja[l]]);*/
          /* w[nz] = -1.;*/ /* use unit weight for now, later can try
                               1/(deg(i)+deg(k)) */
 
          w[nz] = -1 / (dist * dist);
 
          diag_w += w[nz];
 
          jd[nz] = ja[l];
          /*
            d[nz] = w[nz]*(distance(x,dim,i,k)+distance(x,dim,k,ja[l]));
            d[nz] = w[nz]*(dd[j]+dd[l]);
            d[nz] = w[nz]*(avg_dist[i] + 2*avg_dist[k] + avg_dist[ja[l]])*0.5;
          */
          d[nz] = w[nz] * dist;
          stop += d[nz] * distance(x, dim, ja[l], k);
          sbot += d[nz] * dist;
          diag_d += d[nz];
 
          nz++;
        }
      }
    }
    jw[nz] = i;
    lambda[i] *= (-diag_w); /* alternatively don't do that then we have a
                               constant penalty term scaled by lambda0 */
 
    w[nz] = -diag_w + lambda[i];
    jd[nz] = i;
    d[nz] = -diag_d;
    nz++;
 
    assert(nz <= INT_MAX);
    iw[i + 1] = (int)nz;
    id[i + 1] = (int)nz;
  }
  s = stop / sbot;
  for (size_t i = 0; i < nz; i++)
    d[i] *= s;
 
  sm->scaling = s;
  sm->Lw->nz = nz;
  sm->Lwd->nz = nz;
 
  free(mask);
  free(avg_dist);
  SparseMatrix_delete(ID);
  return sm;
}
 
StressMajorizationSmoother
SparseStressMajorizationSmoother_new(SparseMatrix A, int dim, double *x) {
  /* solve a stress model to achieve the ideal distance among a sparse set of
     edges recorded in A. A must be a real matrix.
   */
  int m = A->m;
  double stop = 0, sbot = 0;
 
  assert(SparseMatrix_is_symmetric(A, false) && A->type == MATRIX_TYPE_REAL);
 
  /* if x is all zero, make it random */
  bool has_nonzero = false;
  for (int i = 0; i < m * dim; i++) {
    if (!is_exactly_equal(x[i], 0)) {
      has_nonzero = true;
      break;
    }
  }
  if (!has_nonzero) {
    for (int i = 0; i < m * dim; i++)
      x[i] = 72 * drand();
  }
 
  const int *const ia = A->ia;
  const int *const ja = A->ja;
  const double *const a = A->a;
 
  StressMajorizationSmoother sm =
      gv_alloc(sizeof(struct StressMajorizationSmoother_struct));
  sm->scaling = 1.;
  sm->data = NULL;
  sm->scheme = SM_SCHEME_NORMAL;
  sm->D = A;
  sm->tol_cg = 0.01;
  sm->maxit_cg = floor(sqrt(A->m));
 
  double *const lambda = sm->lambda = gv_calloc(m, sizeof(double));
 
  size_t nz = A->nz;
 
  sm->Lw = SparseMatrix_new(m, m, nz + (size_t)m, MATRIX_TYPE_REAL, FORMAT_CSR);
  assert(sm->Lw != NULL);
  sm->Lwd =
      SparseMatrix_new(m, m, nz + (size_t)m, MATRIX_TYPE_REAL, FORMAT_CSR);
  assert(sm->Lwd != NULL);
 
  int *const iw = sm->Lw->ia;
  int *const jw = sm->Lw->ja;
  int *const id = sm->Lwd->ia;
  int *const jd = sm->Lwd->ja;
  double *const w = sm->Lw->a;
  double *const d = sm->Lwd->a;
  iw[0] = id[0] = 0;
 
  nz = 0;
  for (int i = 0; i < m; i++) {
    double diag_d = 0;
    double diag_w = 0;
    for (int j = ia[i]; j < ia[i + 1]; j++) {
      const int k = ja[j];
      if (k != i) {
 
        jw[nz] = k;
        const double dist = a[j];
        w[nz] = -1;
        diag_w += w[nz];
        jd[nz] = k;
        d[nz] = w[nz] * dist;
 
        stop += d[nz] * distance(x, dim, i, k);
        sbot += d[nz] * dist;
        diag_d += d[nz];
 
        nz++;
      }
    }
 
    jw[nz] = i;
    lambda[i] *= (-diag_w); /* alternatively don't do that then we have a
                               constant penalty term scaled by lambda0 */
    w[nz] = -diag_w + lambda[i];
 
    jd[nz] = i;
    d[nz] = -diag_d;
    nz++;
 
    assert(nz <= INT_MAX);
    iw[i + 1] = (int)nz;
    id[i + 1] = (int)nz;
  }
  const double s = stop / sbot;
  if (s == 0) {
    StressMajorizationSmoother_delete(sm);
    return NULL;
  }
  for (size_t i = 0; i < nz; i++)
    d[i] *= s;
 
  sm->scaling = s;
  sm->Lw->nz = nz;
  sm->Lwd->nz = nz;
 
  return sm;
}
 
static double total_distance(int m, int dim, double *x, double *y) {
  double total = 0, dist = 0;
  int i, j;
 
  for (i = 0; i < m; i++) {
    dist = 0.;
    for (j = 0; j < dim; j++) {
      dist +=
          (y[i * dim + j] - x[i * dim + j]) * (y[i * dim + j] - x[i * dim + j]);
    }
    total += sqrt(dist);
  }
  return total;
}
 
void SparseStressMajorizationSmoother_delete(
    SparseStressMajorizationSmoother sm) {
  StressMajorizationSmoother_delete(sm);
}
 
double
SparseStressMajorizationSmoother_smooth(SparseStressMajorizationSmoother sm,
                                        int dim, double *x, int maxit_sm) {
  return StressMajorizationSmoother_smooth(sm, dim, x, maxit_sm);
}
 
static void get_edge_label_matrix(relative_position_constraints data, int m,
                                  int dim, double *x, SparseMatrix *LL,
                                  double **rhs) {
  int edge_labeling_scheme = data->edge_labeling_scheme;
  int n_constr_nodes = data->n_constr_nodes;
  int *constr_nodes = data->constr_nodes;
  SparseMatrix A_constr = data->A_constr;
  int *ia = A_constr->ia, *ja = A_constr->ja, ii, jj, l, ll, i, j;
  int *irn = data->irn, *jcn = data->jcn;
  double *val = data->val, dist, kk, k;
  double *x00 = NULL;
  SparseMatrix Lc = NULL;
  double constr_penalty = data->constr_penalty;
 
  if (edge_labeling_scheme == ELSCHEME_PENALTY ||
      edge_labeling_scheme == ELSCHEME_STRAIGHTLINE_PENALTY) {
    /* for an node with two neighbors j--i--k, and assume i needs to be between
       j and k, then the contribution to P is .   i    j     k i   1   -0.5 -0.5
       j  -0.5 0.25  0.25
       k  -0.5 0.25  0.25
       in general, if i has m neighbors j, k, ..., then
       pᵢᵢ = 1
       pⱼⱼ = 1 ÷ m²
       pᵢⱼ = -1 ÷ m
       pⱼₖ = 1 ÷ m²
       .    i       j        k
       i    1       -1 ÷ m   -1 ÷ m ...
       j   -1 ÷ m   1 ÷ m²   1 ÷ m² ...
       k   -1 ÷ m   1 ÷ m²   1 ÷ m² ...
    */
    if (!irn) {
      assert((!jcn) && (!val));
      size_t nz = 0;
      for (i = 0; i < n_constr_nodes; i++) {
        ii = constr_nodes[i];
        k = ia[ii + 1] - ia[ii]; /*usually k = 2 */
        nz += (size_t)((k + 1) * (k + 1));
      }
      irn = data->irn = gv_calloc(nz, sizeof(int));
      jcn = data->jcn = gv_calloc(nz, sizeof(int));
      val = data->val = gv_calloc(nz, sizeof(double));
    }
    size_t nz = 0;
    for (i = 0; i < n_constr_nodes; i++) {
      ii = constr_nodes[i];
      jj = ja[ia[ii]];
      ll = ja[ia[ii] + 1];
      if (jj == ll)
        continue; /* do not do loops */
      dist = distance_cropped(x, dim, jj, ll);
      dist *= dist;
 
      k = ia[ii + 1] - ia[ii]; /* usually k = 2 */
      kk = k * k;
      irn[nz] = ii;
      jcn[nz] = ii;
      val[nz++] = constr_penalty / (dist);
      k = constr_penalty / (k * dist);
      kk = constr_penalty / (kk * dist);
      for (j = ia[ii]; j < ia[ii + 1]; j++) {
        irn[nz] = ii;
        jcn[nz] = ja[j];
        val[nz++] = -k;
      }
      for (j = ia[ii]; j < ia[ii + 1]; j++) {
        jj = ja[j];
        irn[nz] = jj;
        jcn[nz] = ii;
        val[nz++] = -k;
        for (l = ia[ii]; l < ia[ii + 1]; l++) {
          ll = ja[l];
          irn[nz] = jj;
          jcn[nz] = ll;
          val[nz++] = kk;
        }
      }
    }
    Lc = SparseMatrix_from_coordinate_arrays(nz, m, m, irn, jcn, val,
                                             MATRIX_TYPE_REAL, sizeof(double));
  } else if (edge_labeling_scheme == ELSCHEME_PENALTY2 ||
             edge_labeling_scheme == ELSCHEME_STRAIGHTLINE_PENALTY2) {
    /* for an node with two neighbors j--i--k, and assume i needs to be between
       the old position of j and k, then the contribution to P is 1/d_jk, and to
       the right hand side: {0,...,average_position_of_i's neighbor if i is an
       edge node,...}
    */
    if (!irn) {
      assert((!jcn) && (!val));
      int nz = n_constr_nodes;
      irn = data->irn = gv_calloc(nz, sizeof(int));
      jcn = data->jcn = gv_calloc(nz, sizeof(int));
      val = data->val = gv_calloc(nz, sizeof(double));
    }
    x00 = gv_calloc(m * dim, sizeof(double));
    size_t nz = 0;
    for (i = 0; i < n_constr_nodes; i++) {
      ii = constr_nodes[i];
      jj = ja[ia[ii]];
      ll = ja[ia[ii] + 1];
      dist = distance_cropped(x, dim, jj, ll);
      irn[nz] = ii;
      jcn[nz] = ii;
      val[nz++] = constr_penalty / (dist);
      for (j = ia[ii]; j < ia[ii + 1]; j++) {
        jj = ja[j];
        for (l = 0; l < dim; l++) {
          x00[ii * dim + l] += x[jj * dim + l];
        }
      }
      for (l = 0; l < dim; l++) {
        x00[ii * dim + l] *= constr_penalty / (dist) / (ia[ii + 1] - ia[ii]);
      }
    }
    Lc = SparseMatrix_from_coordinate_arrays(nz, m, m, irn, jcn, val,
                                             MATRIX_TYPE_REAL, sizeof(double));
  }
  *LL = Lc;
  *rhs = x00;
}
 
static UNUSED double get_stress(int m, int dim, int *iw, int *jw, double *w,
                                double *d, double *x, double scaling) {
  int i, j;
  double res = 0., dist;
  // We use the fact that dᵢⱼ = wᵢⱼ × graph_dist(i, j). Also, dᵢⱼ and x are
  // scaled by ×scaling, so divide by it to get actual unscaled stress.
  for (i = 0; i < m; i++) {
    for (j = iw[i]; j < iw[i + 1]; j++) {
      if (i == jw[j]) {
        continue;
      }
      dist = d[j] / w[j]; /* both negative*/
      res += -w[j] * (dist - distance(x, dim, i, jw[j])) *
             (dist - distance(x, dim, i, jw[j]));
    }
  }
  return 0.5 * res / scaling / scaling;
}
 
double StressMajorizationSmoother_smooth(StressMajorizationSmoother sm, int dim,
                                         double *x, int maxit_sm) {
  SparseMatrix Lw = sm->Lw, Lwd = sm->Lwd, Lwdd = NULL;
  int i, j, k, m, *id, *jd, *iw, *jw, idiag, iter = 0;
  double *w, *dd, *d, *y = NULL, *x0 = NULL, *x00 = NULL, diag, diff = 1,
                      *lambda = sm->lambda;
  SparseMatrix Lc = NULL;
  double dij, dist;
 
  const double tol = 0.001;
 
  Lwdd = SparseMatrix_copy(Lwd);
  m = Lw->m;
  x0 = calloc(dim * m, sizeof(double));
  if (!x0)
    goto RETURN;
 
  memcpy(x0, x, sizeof(double) * dim * m);
  y = calloc(dim * m, sizeof(double));
  if (!y)
    goto RETURN;
 
  id = Lwd->ia;
  jd = Lwd->ja;
  d = Lwd->a;
  dd = Lwdd->a;
  w = Lw->a;
  iw = Lw->ia;
  jw = Lw->ja;
 
#ifdef DEBUG_PRINT
  if (Verbose)
    fprintf(stderr, "initial stress = %f\n",
            get_stress(m, dim, iw, jw, w, d, x, sm->scaling));
#else
  (void)iw;
  (void)jw;
#endif
  /* for the additional matrix L due to the position constraints */
  if (sm->scheme == SM_SCHEME_NORMAL_ELABEL) {
    get_edge_label_matrix(sm->data, m, dim, x, &Lc, &x00);
    if (Lc)
      Lw = SparseMatrix_add(Lw, Lc);
  }
 
  while (iter++ < maxit_sm && diff > tol) {
 
    for (i = 0; i < m; i++) {
      idiag = -1;
      diag = 0.;
      for (j = id[i]; j < id[i + 1]; j++) {
        if (i == jd[j]) {
          idiag = j;
          continue;
        }
 
        dist = distance(x, dim, i, jd[j]);
        if (d[j] == 0) {
          dd[j] = 0;
        } else {
          if (dist == 0) {
            dij = d[j] / w[j]; /* the ideal distance */
            /* perturb so points do not sit at the same place */
            for (k = 0; k < dim; k++)
              x[jd[j] * dim + k] += 0.0001 * (drand() + .0001) * dij;
            dist = distance(x, dim, i, jd[j]);
          }
          dd[j] = d[j] / dist;
        }
        diag += dd[j];
      }
      assert(idiag >= 0);
      dd[idiag] = -diag;
    }
    /* solve (Lw+lambda*I) x = Lwdd y + lambda x0 */
 
    SparseMatrix_multiply_dense(Lwdd, x, y, dim);
 
    if (lambda) { /* is there a penalty term? */
      for (i = 0; i < m; i++) {
        for (j = 0; j < dim; j++) {
          y[i * dim + j] += lambda[i] * x0[i * dim + j];
        }
      }
    }
 
    /* additional term added to the rhs */
    switch (sm->scheme) {
    case SM_SCHEME_NORMAL_ELABEL: {
      for (i = 0; i < m; i++) {
        for (j = 0; j < dim; j++) {
          y[i * dim + j] += x00[i * dim + j];
        }
      }
      break;
    }
    default:
      break;
    }
 
#ifdef DEBUG_PRINT
    if (Verbose) {
      fprintf(stderr, "stress1 = %g\n",
              get_stress(m, dim, iw, jw, w, d, x, sm->scaling));
    }
#endif
 
    SparseMatrix_solve(Lw, dim, x, y, sm->tol_cg, sm->maxit_cg);
 
#ifdef DEBUG_PRINT
    if (Verbose)
      fprintf(stderr, "stress2 = %g\n",
              get_stress(m, dim, iw, jw, w, d, y, sm->scaling));
#endif
    diff = total_distance(m, dim, x, y) / sqrt(vector_product(m * dim, x, x));
#ifdef DEBUG_PRINT
    if (Verbose) {
      fprintf(stderr,
              "Outer iter = %d, cg res = %g, ||x_{k+1}-x_k||/||x_k|| = %g\n",
              iter, res, diff);
    }
#endif
 
    memcpy(x, y, sizeof(double) * m * dim);
  }
 
#ifdef DEBUG
  _statistics[1] += iter - 1;
#endif
 
#ifdef DEBUG_PRINT
  if (Verbose)
    fprintf(stderr, "iter = %d, final stress = %f\n", iter,
            get_stress(m, dim, iw, jw, w, d, x, sm->scaling));
#endif
 
RETURN:
  SparseMatrix_delete(Lwdd);
  if (Lc) {
    SparseMatrix_delete(Lc);
    SparseMatrix_delete(Lw);
  }
 
  free(x0);
  free(y);
  free(x00);
  return diff;
}
 
void StressMajorizationSmoother_delete(StressMajorizationSmoother sm) {
  if (!sm)
    return;
  if (sm->Lw)
    SparseMatrix_delete(sm->Lw);
  if (sm->Lwd)
    SparseMatrix_delete(sm->Lwd);
  free(sm->lambda);
  if (sm->data)
    sm->data_deallocator(sm->data);
  free(sm);
}
 
TriangleSmoother TriangleSmoother_new(SparseMatrix A, int dim, double *x,
                                      bool use_triangularization) {
  int i, j, k, m = A->m, *ia = A->ia, *ja = A->ja, *iw, *jw, jdiag, nz;
  SparseMatrix B;
  double *d, *w, diag_d, diag_w, dist;
  double s = 0, stop = 0, sbot = 0;
 
  assert(SparseMatrix_is_symmetric(A, false));
 
  double *avg_dist = gv_calloc(m, sizeof(double));
 
  for (i = 0; i < m; i++) {
    avg_dist[i] = 0;
    nz = 0;
    for (j = ia[i]; j < ia[i + 1]; j++) {
      if (i == ja[j])
        continue;
      avg_dist[i] += distance(x, dim, i, ja[j]);
      nz++;
    }
    assert(nz > 0);
    avg_dist[i] /= nz;
  }
 
  TriangleSmoother sm = gv_alloc(sizeof(struct TriangleSmoother_struct));
  sm->scaling = 1;
  sm->data = NULL;
  sm->scheme = SM_SCHEME_NORMAL;
  sm->tol_cg = 0.01;
  sm->maxit_cg = floor(sqrt(A->m));
 
  double *lambda = sm->lambda = gv_calloc(m, sizeof(double));
 
  if (m > 2) {
    if (use_triangularization) {
      B = call_tri(m, x);
    } else {
      B = call_tri2(m, dim, x);
    }
  } else {
    B = SparseMatrix_copy(A);
  }
 
  sm->Lw = SparseMatrix_add(A, B);
 
  SparseMatrix_delete(B);
  sm->Lwd = SparseMatrix_copy(sm->Lw);
  if (!sm->Lw || !sm->Lwd) {
    TriangleSmoother_delete(sm);
    return NULL;
  }
 
  iw = sm->Lw->ia;
  jw = sm->Lw->ja;
 
  w = sm->Lw->a;
  d = sm->Lwd->a;
 
  for (i = 0; i < m; i++) {
    diag_d = diag_w = 0;
    jdiag = -1;
    for (j = iw[i]; j < iw[i + 1]; j++) {
      k = jw[j];
      if (k == i) {
        jdiag = j;
        continue;
      }
      /*      w[j] = -1./(ia[i+1]-ia[i]+ia[ja[j]+1]-ia[ja[j]]);
              w[j] = -2./(avg_dist[i]+avg_dist[k]);
              w[j] = -1.*/
      ; /* use unit weight for now, later can try 1/(deg(i)+deg(k)) */
      dist = pow(distance_cropped(x, dim, i, k), 0.6);
      w[j] = 1 / (dist * dist);
      diag_w += w[j];
 
      /*      d[j] = w[j]*distance(x,dim,i,k);
              d[j] = w[j]*(avg_dist[i] + avg_dist[k])*0.5;*/
      d[j] = w[j] * dist;
      stop += d[j] * distance(x, dim, i, k);
      sbot += d[j] * dist;
      diag_d += d[j];
    }
 
    lambda[i] *= (-diag_w);
 
    assert(jdiag >= 0);
    w[jdiag] = -diag_w + lambda[i];
    d[jdiag] = -diag_d;
  }
 
  s = stop / sbot;
  for (i = 0; i < iw[m]; i++)
    d[i] *= s;
  sm->scaling = s;
 
  free(avg_dist);
 
  return sm;
}
 
void TriangleSmoother_delete(TriangleSmoother sm) {
 
  StressMajorizationSmoother_delete(sm);
}
 
void TriangleSmoother_smooth(TriangleSmoother sm, int dim, double *x) {
  StressMajorizationSmoother_smooth(sm, dim, x, 50);
}
 
/* ================================ spring and spring-electrical based smoother
 * ================ */
SpringSmoother SpringSmoother_new(SparseMatrix A, int dim,
                                  spring_electrical_control ctrl, double *x) {
  int j, k, l, m = A->m, *ia = A->ia, *ja = A->ja, *id, *jd;
  double *d, *dd;
  SparseMatrix ID = NULL;
 
  assert(SparseMatrix_is_symmetric(A, false));
 
  ID = ideal_distance_matrix(A, dim, x);
  dd = ID->a;
 
  SpringSmoother sm = gv_alloc(sizeof(struct SpringSmoother_struct));
  int *mask = gv_calloc(m, sizeof(int));
 
  double *avg_dist = gv_calloc(m, sizeof(double));
 
  for (int i = 0; i < m; i++) {
    avg_dist[i] = 0;
    int nz = 0;
    for (j = ia[i]; j < ia[i + 1]; j++) {
      if (i == ja[j])
        continue;
      avg_dist[i] += distance(x, dim, i, ja[j]);
      nz++;
    }
    assert(nz > 0);
    avg_dist[i] /= nz;
  }
 
  for (int i = 0; i < m; i++)
    mask[i] = -1;
 
  size_t nz = 0;
  for (int i = 0; i < m; i++) {
    mask[i] = i;
    for (j = ia[i]; j < ia[i + 1]; j++) {
      k = ja[j];
      if (mask[k] != i) {
        mask[k] = i;
        nz++;
      }
    }
    for (j = ia[i]; j < ia[i + 1]; j++) {
      k = ja[j];
      for (l = ia[k]; l < ia[k + 1]; l++) {
        if (mask[ja[l]] != i) {
          mask[ja[l]] = i;
          nz++;
        }
      }
    }
  }
 
  sm->D = SparseMatrix_new(m, m, nz, MATRIX_TYPE_REAL, FORMAT_CSR);
  assert(sm->D != NULL);
 
  id = sm->D->ia;
  jd = sm->D->ja;
  d = sm->D->a;
  id[0] = 0;
 
  nz = 0;
  for (int i = 0; i < m; i++) {
    mask[i] = i + m;
    for (j = ia[i]; j < ia[i + 1]; j++) {
      k = ja[j];
      if (mask[k] != i + m) {
        mask[k] = i + m;
        jd[nz] = k;
        d[nz] = (avg_dist[i] + avg_dist[k]) * 0.5;
        d[nz] = dd[j];
        nz++;
      }
    }
 
    for (j = ia[i]; j < ia[i + 1]; j++) {
      k = ja[j];
      for (l = ia[k]; l < ia[k + 1]; l++) {
        if (mask[ja[l]] != i + m) {
          mask[ja[l]] = i + m;
          jd[nz] = ja[l];
          d[nz] = (avg_dist[i] + 2 * avg_dist[k] + avg_dist[ja[l]]) * 0.5;
          d[nz] = dd[j] + dd[l];
          nz++;
        }
      }
    }
    assert(nz <= INT_MAX);
    id[i + 1] = (int)nz;
  }
  sm->D->nz = nz;
  sm->ctrl = ctrl;
  sm->ctrl.random_start = false;
  sm->ctrl.multilevels = 1;
  sm->ctrl.step /= 2;
  sm->ctrl.maxiter = 20;
 
  free(mask);
  free(avg_dist);
  SparseMatrix_delete(ID);
 
  return sm;
}
 
void SpringSmoother_delete(SpringSmoother sm) {
  if (!sm)
    return;
  if (sm->D)
    SparseMatrix_delete(sm->D);
}
 
void SpringSmoother_smooth(SpringSmoother sm, SparseMatrix A, int dim,
                           double *x) {
  int flag = 0;
 
  spring_electrical_spring_embedding(dim, A, sm->D, &sm->ctrl, x, &flag);
  assert(!flag);
}
 
/*=============================== end of spring and spring-electrical based
 * smoother =========== */
 
void post_process_smoothing(int dim, SparseMatrix A,
                            spring_electrical_control ctrl, double *x) {
#ifdef TIME
  clock_t cpu;
#endif
 
#ifdef TIME
  cpu = clock();
#endif
 
  switch (ctrl.smoothing) {
  case SMOOTHING_RNG:
  case SMOOTHING_TRIANGLE: {
    TriangleSmoother sm;
 
    if (A->m > 2) { /* triangles need at least 3 nodes */
      if (ctrl.smoothing == SMOOTHING_RNG) {
        sm = TriangleSmoother_new(A, dim, x, false);
      } else {
        sm = TriangleSmoother_new(A, dim, x, true);
      }
      TriangleSmoother_smooth(sm, dim, x);
      TriangleSmoother_delete(sm);
    }
    break;
  }
  case SMOOTHING_STRESS_MAJORIZATION_GRAPH_DIST:
  case SMOOTHING_STRESS_MAJORIZATION_POWER_DIST:
  case SMOOTHING_STRESS_MAJORIZATION_AVG_DIST: {
    StressMajorizationSmoother sm;
    int dist_scheme = IDEAL_AVG_DIST;
 
    if (ctrl.smoothing == SMOOTHING_STRESS_MAJORIZATION_GRAPH_DIST) {
      dist_scheme = IDEAL_GRAPH_DIST;
    } else if (ctrl.smoothing == SMOOTHING_STRESS_MAJORIZATION_AVG_DIST) {
      dist_scheme = IDEAL_AVG_DIST;
    } else if (ctrl.smoothing == SMOOTHING_STRESS_MAJORIZATION_POWER_DIST) {
      dist_scheme = IDEAL_POWER_DIST;
    }
 
    sm = StressMajorizationSmoother2_new(A, dim, 0.05, x, dist_scheme);
    StressMajorizationSmoother_smooth(sm, dim, x, 50);
    StressMajorizationSmoother_delete(sm);
    break;
  }
  case SMOOTHING_SPRING: {
    SpringSmoother sm = SpringSmoother_new(A, dim, ctrl, x);
    SpringSmoother_smooth(sm, A, dim, x);
    SpringSmoother_delete(sm);
 
    break;
  }
 
  } /* end switch between smoothing methods */
 
#ifdef TIME
  if (Verbose)
    fprintf(stderr, "post processing %f\n",
            ((double)(clock() - cpu)) / CLOCKS_PER_SEC);
#endif
}