TrajectoryDataset

A Dataset specialized for trajectory data.

Constructor

TrajectoryDataset.from_attributes(position, ...)
TrajectoryDataset.from_trajectories(tjs)	Create a dataset based on trajectories

Serialization / IO / conversion

TrajectoryDataset.from_csv(file[, indices])	Read a dictionary-alike object from a .csv file as a pandas DataFrame.
TrajectoryDataset.as_geo(*args[, activate])	Convert to a geospatial representation using geopandas.

Indexing

TrajectoryDataset.att(timestamps[, columns])
TrajectoryDataset.atid(ids[, attributes])	Select rows by the given id and optionally by attributes
TrajectoryDataset.trajectory(index[, inverse])	Select trajectory data for specific indices, or exclude them if inverse is set to True.

Filtering

TrajectoryDataset.get_by_object_class(...)	Return only the poses of a specific object class.
TrajectoryDataset.during(since, until[, ...])	Select rows within a specific time range (include "since", exclude "until").

Analysis

TrajectoryDataset.most_likely_class([by, ...])	Get the name of the most probable object class for each pose or trajectory of the dataset
TrajectoryDataset.roi	Return the region of interest of the dataset.

Attributes

TrajectoryDataset.attributes	Returns the dataset attributes
TrajectoryDataset.ids	Returns the unique ids in the dataset
TrajectoryDataset.interval	Returns the time interval this object spans
TrajectoryDataset.timestamps	The unique timestamps in the dataframe

Note

All pandas DataFrame methods are also available.