dsf::mdt::TrajectoryCollection Class Reference

Public Member Functions
	TrajectoryCollection (std::unordered_map< std::string, std::variant< std::vector< Id >, std::vector< std::time_t >, std::vector< double > > > &&dataframe, std::array< double, 4 > const &bbox={})
	Construct a TrajectoryCollection from a dataframe.
	TrajectoryCollection (std::string const &fileName=std::string(), std::unordered_map< std::string, std::string > const &column_mapping={}, char const sep=';', std::array< double, 4 > const &bbox={})
	Construct a TrajectoryCollection, optionally importing from a CSV file.
void	import (std::string const &fileName, std::unordered_map< std::string, std::string > const &column_mapping={}, char const sep=';', std::array< double, 4 > const &bbox={})
	Import trajectories from a CSV file.
void	to_csv (std::string const &fileName, char const sep=';') const
	Export clustered trajectories to a CSV file with columns 'uid', 'trajectory_id', 'lon', 'lat', 'timestamp_in', 'timestamp_out'.
void	filter (double const cluster_radius_km, double const max_speed_kph=150.0, std::size_t const min_points_per_trajectory=2, std::optional< std::time_t > const min_duration_min=std::nullopt)
	Filter all point trajectories to identify stop points based on clustering and speed criteria.
auto const &	trajectories () const noexcept
	Get the underlying trajectories map.

Constructor & Destructor Documentation

TrajectoryCollection() [1/2]

dsf::mdt::TrajectoryCollection::TrajectoryCollection	(	std::unordered_map< std::string, std::variant< std::vector< Id >, std::vector< std::time_t >, std::vector< double > > > &&	dataframe,
		std::array< double, 4 > const &	bbox = {} )

Construct a TrajectoryCollection from a dataframe.

Parameters

dataframe	An unordered_map representing the dataframe with columns: 'uid', 'timestamp', 'lat', 'lon'.
bbox	Optional bounding box [minX, minY, maxX, maxY] to limit the area of interest. Default is empty (no bounding box).

TrajectoryCollection() [2/2]

dsf::mdt::TrajectoryCollection::TrajectoryCollection	(	std::string const &	fileName = std::string(),
		std::unordered_map< std::string, std::string > const &	column_mapping = {},
		char const	sep = ';',
		std::array< double, 4 > const &	bbox = {} )

Construct a TrajectoryCollection, optionally importing from a CSV file.

Parameters

fileName	The path to the CSV file.
column_mapping	A mapping of column names.
sep	The character used to separate values in the CSV file.
bbox	Optional bounding box [minX, minY, maxX, maxY] to limit the area of interest. Default is empty (no bounding box).

Member Function Documentation

filter()

void dsf::mdt::TrajectoryCollection::filter	(	double const	cluster_radius_km,
		double const	max_speed_kph = 150.0,
		std::size_t const	min_points_per_trajectory = 2,
		std::optional< std::time_t > const	min_duration_min = std::nullopt )

Filter all point trajectories to identify stop points based on clustering and speed criteria.

Parameters

cluster_radius_km	The radius (in kilometers) to use for clustering points.
max_speed_kph	The max allowed speed (in km/h) to consider a cluster as a stop point. Default is 150.0 km/h.
min_points_per_trajectory	The minimum number of points required for a trajectory to be considered valid. Default is 2.
min_duration_min	The minimum duration (in minutes) for a cluster to be considered a stop point. If stops are detected, trajectories may be split into multiple segments.

import()

void dsf::mdt::TrajectoryCollection::import	(	std::string const &	fileName,
		std::unordered_map< std::string, std::string > const &	column_mapping = {},
		char const	sep = ';',
		std::array< double, 4 > const &	bbox = {} )

Import trajectories from a CSV file.

Parameters

fileName	The path to the CSV file. Accepts columns: 'uid', 'timestamp', 'lat', 'lon'.
column_mapping	A mapping of column names.
sep	The character used to separate values in the CSV file.
bbox	Optional bounding box [minX, minY, maxX, maxY] to limit the area of interest. Default is empty (no bounding box).

to_csv()

void dsf::mdt::TrajectoryCollection::to_csv	(	std::string const &	fileName,
		char const	sep = ';' ) const

Export clustered trajectories to a CSV file with columns 'uid', 'trajectory_id', 'lon', 'lat', 'timestamp_in', 'timestamp_out'.

Parameters

fileName	The path to the output CSV file.
sep	The character used to separate values in the CSV file.

trajectories()

auto const & dsf::mdt::TrajectoryCollection::trajectories ( ) const

inlinenoexcept

Get the underlying trajectories map.

Returns

A const reference to the map of trajectories.

---

The documentation for this class was generated from the following files:

• src/dsf/mdt/TrajectoryCollection.hpp
• src/dsf/mdt/TrajectoryCollection.cpp