Open sglvladi opened 1 year ago
Hey, I don't know if it is my implementation that it is not working fine or something more complex, but if I use a JPDA data associator in a simple Particle filter (so no ELPF) I get this error:
ValueError: Length of weights not compatible with specified axis.
A "not working" example can be found here, if in the example you switch to a distance based data associator it is working fine.
Attention: Patch coverage is 94.50549%
with 5 lines
in your changes are missing coverage. Please review.
Project coverage is 93.52%. Comparing base (
8c8f9ee
) to head (4fe0fc0
).
Files | Patch % | Lines |
---|---|---|
stonesoup/tracker/particle.py | 94.50% | 2 Missing and 3 partials :warning: |
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
This PR adds the following Tracker classes:
SingleTargetExpectedLikelihoodParticleFilter
MultiTargetExpectedLikelihoodParticleFilter
The above are single/multi target implementations of the approach described in [1]. Essentially, these implement (J)PDA for particle filtering, but can also work with (G)NN data associators.
A running example can be found here.
[1] Marrs, A., Maskell, S., and Bar-Shalom, Y., “Expected likelihood for tracking in clutter with particle filters”, in Signal and Data Processing of Small Targets 2002, 2002, vol. 4728, pp. 230–239. doi:10.1117/12.478507.