jubatus::core::clustering Namespace Reference

Namespaces
	compressor

Classes
class	clustering
struct	clustering_config
class	clustering_error
class	clustering_method
class	clustering_method_factory
class	compressive_storage
class	discrete_distribution
class	eigen_feature_mapper
struct	eigen_wsvec_t
class	event_dispatcher
class	gmm
class	gmm_clustering_method
class	kmeans_clustering_method
class	simple_storage
class	storage
class	storage_factory
struct	weighted_point

Typedefs
typedef std::vector< cluster_unit >	cluster_set
typedef std::vector< std::pair< cluster_weight, jubatus::core::fv_converter::datum > >	cluster_unit
typedef double	cluster_weight
typedef std::vector< std::pair< std::string, wplist > >	diff_t
typedef std::vector< eigen_smat_t >	eigen_smat_list_t
typedef Eigen::SparseMatrix< double >	eigen_smat_t
typedef std::vector< jubatus::util::lang::shared_ptr< eigen_solver_t > >	eigen_solver_list_t
typedef Eigen::SimplicialCholesky< eigen_smat_t >	eigen_solver_t
typedef std::vector< eigen_svec_t >	eigen_svec_list_t
typedef Eigen::SparseVector< double >	eigen_svec_t
typedef std::vector< eigen_wsvec_t >	eigen_wsvec_list_t
typedef framework::linear_mixable_helper< storage, diff_t >	mixable_storage
typedef std::vector< common::sfv_t >	plist
typedef std::vector< weighted_point >	wplist

Enumerations
enum	storage_event_type { REVISION_CHANGE, UPDATE }

Functions
common::sfv_t	add (const common::sfv_t &p1, const common::sfv_t &p2)
void	concat (const wplist &src, wplist &dst)
char	digit (int num, int r, int n)
double	dist (const common::sfv_t &p1, const common::sfv_t &p2)
double	dist (const weighted_point &d1, const weighted_point &d2)
void	dump_wplist (const wplist &src)
weighted_point	get_point (size_t d)
weighted_point	get_point (size_t d, const std::vector< double > &c)
std::vector< weighted_point >	get_points (size_t n, size_t d)
std::vector< weighted_point >	get_points (size_t n, size_t d, const std::vector< double > &c)
std::pair< size_t, double >	min_dist (const common::sfv_t &p, const std::vector< common::sfv_t > &P)
pair< size_t, double >	min_dist (const common::sfv_t &p, const vector< common::sfv_t > &P)
std::pair< size_t, double >	min_dist (const weighted_point &d1, const wplist &P)
common::sfv_t	scalar_dot (const common::sfv_t &p, double s)
void	scalar_mul_and_add (const common::sfv_t &left, float s, common::sfv_t &right)
common::sfv_t	sub (const common::sfv_t &p1, const common::sfv_t &p2)
double	sum (const common::sfv_t &p)
double	sum2 (const common::sfv_t &p)
void	swap (weighted_point &p1, weighted_point &p2)

Typedef Documentation

typedef std::vector<cluster_unit> jubatus::core::clustering::cluster_set

Definition at line 37 of file types.hpp.

typedef std::vector<std::pair<cluster_weight, jubatus::core::fv_converter::datum> > jubatus::core::clustering::cluster_unit

Definition at line 36 of file types.hpp.

typedef double jubatus::core::clustering::cluster_weight

Definition at line 34 of file types.hpp.

typedef std::vector<std::pair<std::string, wplist> > jubatus::core::clustering::diff_t

Definition at line 57 of file types.hpp.

typedef std::vector<eigen_smat_t> jubatus::core::clustering::eigen_smat_list_t

Definition at line 33 of file gmm_types.hpp.

typedef Eigen::SparseMatrix<double> jubatus::core::clustering::eigen_smat_t

Definition at line 30 of file gmm_types.hpp.

typedef std::vector<jubatus::util::lang::shared_ptr<eigen_solver_t> > jubatus::core::clustering::eigen_solver_list_t

Definition at line 35 of file gmm_types.hpp.

typedef Eigen::SimplicialCholesky<eigen_smat_t> jubatus::core::clustering::eigen_solver_t

Definition at line 31 of file gmm_types.hpp.

typedef std::vector<eigen_svec_t> jubatus::core::clustering::eigen_svec_list_t

Definition at line 32 of file gmm_types.hpp.

typedef Eigen::SparseVector<double> jubatus::core::clustering::eigen_svec_t

Definition at line 29 of file gmm_types.hpp.

typedef std::vector<eigen_wsvec_t> jubatus::core::clustering::eigen_wsvec_list_t

Definition at line 42 of file gmm_types.hpp.

typedef framework::linear_mixable_helper<storage, diff_t> jubatus::core::clustering::mixable_storage

Definition at line 79 of file storage.hpp.

typedef std::vector<common::sfv_t> jubatus::core::clustering::plist

Definition at line 56 of file types.hpp.

typedef std::vector<weighted_point> jubatus::core::clustering::wplist

Definition at line 55 of file types.hpp.

Enumeration Type Documentation

enum jubatus::core::clustering::storage_event_type

Enumerator
REVISION_CHANGE
UPDATE

Definition at line 34 of file storage.hpp.

                        {
  REVISION_CHANGE,
  UPDATE
};

Function Documentation

common::sfv_t jubatus::core::clustering::add	(	const common::sfv_t &	p1,
		const common::sfv_t &	p2
	)

Definition at line 89 of file util.cpp.

                                                              {
  common::sfv_t ret;
  common::sfv_t::const_iterator it1 = p1.begin();
  common::sfv_t::const_iterator it2 = p2.begin();
  while (it1 != p1.end() && it2 != p2.end()) {
    if ((*it1).first < (*it2).first) {
      ret.push_back((*it1));
      ++it1;
    } else if ((*it1).first > (*it2).first) {
      ret.push_back((*it2));
      ++it2;
    } else {
      ret.push_back(make_pair((*it1).first, (*it1).second + (*it2).second));
      ++it1;
      ++it2;
    }
  }
  for (; it1 != p1.end(); ++it1) {
    ret.push_back((*it1));
  }
  for (; it2 != p2.end(); ++it2) {
    ret.push_back((*it2));
  }

  return ret;
}

void jubatus::core::clustering::concat	(	const wplist &	src,
		wplist &	dst
	)

Definition at line 33 of file util.cpp.

Referenced by jubatus::core::clustering::compressive_storage::carry_up(), jubatus::core::clustering::compressor::kmeans_compressor::compress(), jubatus::core::clustering::storage::get_all(), jubatus::core::clustering::storage::get_common(), and jubatus::core::clustering::compressive_storage::get_mine().

                                            {
  dst.insert(dst.end(), src.begin(), src.end());
}

Here is the caller graph for this function:

char jubatus::core::clustering::digit	(	int	num,
		int	r,
		int	n
	)

Definition at line 37 of file util.cpp.

Referenced by jubatus::core::clustering::compressive_storage::is_next_bucket_full().

                                  {
  if (r < 0) {
    return 0;
  }
  for (int i = 0; i < r; ++i) {
    num /= n;
  }
  return num % n;
}

Here is the caller graph for this function:

double jubatus::core::clustering::dist	(	const common::sfv_t &	p1,
		const common::sfv_t &	p2
	)

Definition at line 151 of file util.cpp.

Referenced by dist(), jubatus::core::clustering::kmeans_clustering_method::do_batch_update(), jubatus::core::graph::graph_wo_index::get_diff_eigen_score(), jubatus::core::storage::lsh_index_storage::get_sorted_similar_rows(), jubatus::core::storage::lsh_probe_generator::init(), min_dist(), jubatus::core::nearest_neighbor::ranking_hamming_bit_vectors(), and jubatus::core::graph::graph_wo_index::update_spt_node().

                                                          {
  double ret = 0;
  common::sfv_t::const_iterator it1 = p1.begin();
  common::sfv_t::const_iterator it2 = p2.begin();
  while (it1 != p1.end() && it2 != p2.end()) {
    int cmp = strcmp(it1->first.c_str(), it2->first.c_str());
    if (cmp < 0) {
      ret += it1->second * it1->second;
      ++it1;
    } else if (cmp > 0) {
      ret += it2->second * it2->second;
      ++it2;
    } else {
      ret += (it1->second  - it2->second) * (it1->second - it2->second);
      ++it1;
      ++it2;
    }
  }
  for (; it1 != p1.end(); ++it1) {
    ret += std::pow(it1->second, 2);
  }
  for (; it2 != p2.end(); ++it2) {
    ret += std::pow(it2->second, 2);
  }
  return std::sqrt(ret);
}

Here is the caller graph for this function:

double jubatus::core::clustering::dist	(	const weighted_point &	d1,
		const weighted_point &	d2
	)

Definition at line 178 of file util.cpp.

References jubatus::core::clustering::weighted_point::data, and dist().

                                                                {
  return dist(d1.data, d2.data);
}

Here is the call graph for this function:

void jubatus::core::clustering::dump_wplist

(

const wplist &

src

)

weighted_point jubatus::core::clustering::get_point ( size_t  d )

inline

Definition at line 38 of file testutil.hpp.

References jubatus::core::clustering::weighted_point::data, and jubatus::core::clustering::weighted_point::weight.

Referenced by get_points().

                                          {
  static mtrand r_(0);
  weighted_point ret;
  for (size_t i = 0; i < d; ++i) {
    ret.data.push_back(make_pair(
        lexical_cast<string, int>(i), r_.next_gaussian(0, 10)));
  }
  ret.weight = 1;
  return ret;
}

Here is the caller graph for this function:

weighted_point jubatus::core::clustering::get_point ( size_t  d, const std::vector< double > &  c  )

inline

Definition at line 49 of file testutil.hpp.

References jubatus::core::clustering::weighted_point::data, and jubatus::core::clustering::weighted_point::weight.

                                                                      {
  static mtrand r_(0);
  weighted_point ret;
  for (size_t i = 0; i < d; ++i) {
    ret.data.push_back(std::make_pair(
        "#test_" + lexical_cast<string, int>(i),
        c[i] + r_.next_gaussian(0, 10)));
  }
  ret.weight = 1;
  return ret;
}

std::vector<weighted_point> jubatus::core::clustering::get_points ( size_t  n, size_t  d  )

inline

Definition at line 61 of file testutil.hpp.

References get_point().

                                                                {
  std::vector<weighted_point> ret;
  for (size_t i = 0; i < n; ++i) {
    ret.push_back(get_point(d));
  }
  return ret;
}

Here is the call graph for this function:

std::vector<weighted_point> jubatus::core::clustering::get_points ( size_t  n, size_t  d, const std::vector< double > &  c  )

inline

Definition at line 69 of file testutil.hpp.

References get_point().

                                                                          {
  std::vector<weighted_point> ret;
  for (size_t i = 0; i < n; ++i) {
    ret.push_back(get_point(d, c));
  }
  return ret;
}

Here is the call graph for this function:

std::pair<size_t, double> jubatus::core::clustering::min_dist	(	const common::sfv_t &	p,
		const std::vector< common::sfv_t > &	P
	)

pair<size_t, double> jubatus::core::clustering::min_dist	(	const common::sfv_t &	p,
		const vector< common::sfv_t > &	P
	)

Definition at line 182 of file util.cpp.

References dist().

Referenced by jubatus::core::clustering::compressor::kmeans_compressor::bicriteria_to_coreset(), jubatus::core::clustering::kmeans_clustering_method::do_batch_update(), jubatus::core::clustering::compressor::kmeans_compressor::get_bicriteria(), jubatus::core::clustering::kmeans_clustering_method::get_clusters(), jubatus::core::clustering::kmeans_clustering_method::get_nearest_center_index(), and jubatus::core::clustering::kmeans_clustering_method::initialize_centers().

                                  {
  size_t idx = 0;
  double mindist = DBL_MAX;
  for (vector<common::sfv_t>::const_iterator it = P.begin();
       it != P.end(); ++it) {
    double d = dist(p, *it);
    if (mindist > d) {
      idx = it - P.begin();
      mindist = d;
    }
  }
  return std::make_pair(idx, mindist);
}

Here is the call graph for this function:

Here is the caller graph for this function:

std::pair< size_t, double > jubatus::core::clustering::min_dist	(	const weighted_point &	d1,
		const wplist &	P
	)

Definition at line 198 of file util.cpp.

References dist().

                                                                            {
  double md = DBL_MAX;
  size_t midx = 0;
  for (wplist::const_iterator it = P.begin(); it != P.end(); ++it) {
    double d = dist((*it), d1);
    if (md > d) {
      midx = it - P.begin();
      md = d;
    }
  }
  return std::make_pair(midx, md);
}

Here is the call graph for this function:

common::sfv_t jubatus::core::clustering::scalar_dot	(	const common::sfv_t &	p,
		double	s
	)

Definition at line 143 of file util.cpp.

Referenced by jubatus::core::clustering::kmeans_clustering_method::do_batch_update().

                                                       {
  common::sfv_t ret;
  for (common::sfv_t::const_iterator it = p.begin(); it != p.end(); ++it)  {
    ret.push_back(make_pair((*it).first, (*it).second*s));
  }
  return ret;
}

Here is the caller graph for this function:

void jubatus::core::clustering::scalar_mul_and_add	(	const common::sfv_t &	left,
		float	s,
		common::sfv_t &	right
	)

Definition at line 62 of file util.cpp.

Referenced by jubatus::core::clustering::kmeans_clustering_method::do_batch_update().

                        {
  common::sfv_t::const_iterator l = left.begin();
  common::sfv_t::iterator r = right.begin();
  while (l != left.end() && r != right.end()) {
    if (l->first < r->first) {
      std::pair<std::string, float> p = *l;
      p.second *= s;
      r = right.insert(r, p);
      ++l;
    } else if (l->first > r->first) {
      ++r;
    } else {
      r->second += l->second * s;
      ++l;
      ++r;
    }
  }
  for (; l != left.end(); ++l) {
    std::pair<std::string, float> p = *l;
    p.second *= s;
    right.push_back(p);
  }
}

Here is the caller graph for this function:

common::sfv_t jubatus::core::clustering::sub	(	const common::sfv_t &	p1,
		const common::sfv_t &	p2
	)

Definition at line 116 of file util.cpp.

                                                              {
  common::sfv_t ret;
  common::sfv_t::const_iterator it1 = p1.begin();
  common::sfv_t::const_iterator it2 = p2.begin();
  while (it1 != p1.end() && it2 != p2.end()) {
    if ((*it1).first < (*it2).first) {
      ret.push_back((*it1));
      ++it1;
    } else if ((*it1).first > (*it2).first) {
      ret.push_back(make_pair((*it2).first, -(*it2).second));
      ++it2;
    } else {
      ret.push_back(make_pair((*it1).first, (*it1).second - (*it2).second));
      ++it1;
      ++it2;
    }
  }
  for (; it1 != p1.end(); ++it1) {
    ret.push_back((*it1));
  }
  for (; it2 != p2.end(); ++it2) {
    ret.push_back(make_pair((*it2).first, -(*it2).second));
  }

  return ret;
}

double jubatus::core::clustering::sum

(

const common::sfv_t &

p

)

Definition at line 47 of file util.cpp.

Referenced by jubatus::core::clustering::gmm::cluster_probs(), and jubatus::core::storage::feature_fabssum().

                                 {
  double s = 0;
  for (common::sfv_t::const_iterator it = p.begin(); it != p.end(); ++it) {
    s += (*it).second;
  }
  return s;
}

Here is the caller graph for this function:

double jubatus::core::clustering::sum2

(

const common::sfv_t &

p

)

Definition at line 54 of file util.cpp.

                                  {
  double s = 0;
  for (common::sfv_t::const_iterator it = p.begin(); it != p.end(); ++it) {
    s += std::pow((*it).second, 2);
  }
  return s;
}

void jubatus::core::clustering::swap ( weighted_point &  p1, weighted_point &  p2  )

inline

Definition at line 47 of file types.hpp.

References jubatus::core::clustering::weighted_point::data, jubatus::core::fv_converter::datum::num_values_, jubatus::core::clustering::weighted_point::original, jubatus::core::fv_converter::datum::string_values_, and jubatus::core::clustering::weighted_point::weight.

Referenced by jubatus::core::recommender::lsh::clear(), jubatus::core::fv_converter::counter< std::string >::clear(), jubatus::core::recommender::euclid_lsh::clear(), jubatus::core::anomaly::light_lof::collect_lrds(), jubatus::core::storage::column_table::delete_row_(), jubatus::core::framework::linear_function_mixer::put_diff(), jubatus::core::storage::typed_column< T >::remove(), jubatus::core::graph::graph_wo_index::remove_by_swap(), jubatus::core::burst::basic_window< batch_result >::swap(), jubatus::core::storage::column_type::swap(), and jubatus::core::storage::bit_vector_base< uint64_t >::swap().

                                                         {
  using std::swap;
  swap(p1.weight, p2.weight);
  swap(p1.data, p2.data);
  swap(p1.original.string_values_, p2.original.string_values_);
  swap(p1.original.num_values_, p2.original.num_values_);
}

Here is the caller graph for this function: