22Tracking
33========
44
5+
6+ Default method
7+ **************
8+
59To run a tracking just create an yaml file with minimal specification (*FILES_PATTERN * and *SAVE_DIR *).
610
711Example of yaml
@@ -24,10 +28,45 @@ To run:
2428
2529 EddyTracking conf.yaml -v DEBUG
2630
31+ It will use default tracker:
32+
33+ - No travel longer than 125 km between two observation
34+ - Amplitude and speed radius must be close to previous observation
35+ - In case of several candidate only closest is kept
36+
2737
2838It will produce 4 files by run:
2939
3040- A file of correspondances which will contains all the information to merge all identifications file
3141- A file which will contains all the observations which are alone
3242- A file which will contains all the short track which are shorter than TRACK_DURATION_MIN
33- - A file which will contains all the long track which are longer than TRACK_DURATION_MIN
43+ - A file which will contains all the long track which are longer than TRACK_DURATION_MIN
44+
45+
46+ Choose a tracker
47+ ****************
48+
49+ With yaml you could also select another tracker:
50+
51+ .. code-block :: yaml
52+
53+ PATHS :
54+ # Files produces with EddyIdentification
55+ FILES_PATTERN : MY/IDENTIFICATION_PATH/Anticyclonic*.nc
56+ SAVE_DIR : MY_OUTPUT_PATH
57+
58+ # Number of timestep for missing detection
59+ VIRTUAL_LENGTH_MAX : 3
60+ # Minimal time to consider as a full track
61+ TRACK_DURATION_MIN : 10
62+
63+ CLASS :
64+ # Give the module to import,
65+ # must be available when you do "import module" in python
66+ MODULE : py_eddy_tracker.featured_tracking.old_tracker_reference
67+ # Give class name which must be inherit from
68+ # py_eddy_tracker.observations.observation.EddiesObservations
69+ CLASS : CheltonTracker
70+
71+ This tracker is like described in CHELTON11[https://doi.org/10.1016/j.pocean.2011.01.002].
72+ Code is here :meth: `py_eddy_tracker.featured_tracking.old_tracker_reference `
0 commit comments