pxmlw6n2f/Gazebo_Distributed_MPI/gazebo/math/Kmeans.hh

102 lines
3.7 KiB
C++

/*
* Copyright (C) 2014 Open Source Robotics Foundation
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*
*/
#ifndef _GAZEBO_KMEANS_HH_
#define _GAZEBO_KMEANS_HH_
#include <vector>
#include "gazebo/math/Vector3.hh"
#include "gazebo/util/system.hh"
namespace gazebo
{
namespace math
{
/// \addtogroup gazebo_math
/// \{
/// \class Kmeans Kmeans.hh math/gzmath.hh
/// \brief K-Means clustering algorithm. Given a set of observations,
/// k-means partitions the observations into k sets so as to minimize the
/// within-cluster sum of squares.
/// Description based on http://en.wikipedia.org/wiki/K-means_clustering.
class GZ_MATH_VISIBLE Kmeans
{
/// \brief constructor
/// \param[in] _obs Set of observations to cluster.
public: Kmeans(const std::vector<Vector3> &_obs);
/// \brief Destructor
public: virtual ~Kmeans();
/// \brief Get the observations to cluster.
/// \return The vector of observations.
public: std::vector<Vector3> Observations() const;
/// \brief Set the observations to cluster.
/// \param[in] _obs The new vector of observations.
/// \return True if the vector is not empty or false otherwise.
public: bool Observations(const std::vector<Vector3> &_obs);
/// \brief Add observations to the cluster.
/// \param[in] _obs Vector of observations.
/// \return True if the _obs vector is not empty or false otherwise.
public: bool AppendObservations(const std::vector<Vector3> &_obs);
/// \brief Executes the k-means algorithm.
/// \param[in] _k Number of partitions to cluster.
/// \param[out] _centroids Vector of centroids. Each element contains the
/// centroid of one cluster.
/// \param[out] _labels Vector of labels. The size of this vector is
/// equals to the number of observations. Each element represents the
/// cluster to which observation belongs.
/// \return True when the operation succeed or false otherwise. The
/// operation will fail if the number of observations is not positive,
/// if the number of clusters is non positive, or if the number of
/// clusters if greater than the number of observations.
public: bool Cluster(int _k,
std::vector<Vector3> &_centroids,
std::vector<unsigned int> &_labels);
/// \brief Given an observation, it returns the closest centroid to it.
/// \param[in] _p Point to check.
/// \return The index of the closest centroid to the point _p.
private: unsigned int ClosestCentroid(const Vector3 &_p) const;
/// \brief Observations.
private: std::vector<Vector3> obs;
/// \brief Centroids.
private: std::vector<Vector3> centroids;
/// \brief Each element stores the cluster to which observation i belongs.
private: std::vector<unsigned int> labels;
/// \brief Used to calculate the centroid of each partition.
private: std::vector<Vector3> sums;
/// \brief Counts the number of observations contained in each partition.
private: std::vector<unsigned int> counters;
};
/// \}
}
}
#endif