Steven Hiscocks

Results 35 issues of Steven Hiscocks

Generation of hypotheses for data association could be done (in part) across multiple threads/processors, which should produce performance benefit with many tracks/detections (ideally used in conjunction with tree structures).

enhancement

Since #365, all models `function` methods need to be vectorised. Whilst this was done for particle filter, this may also be useful (minor speed improvement?) on other methods, like the...

good first issue

Should include plugins in general as well as serialisation extensions.

tests

The tree mixins are useful to enable using the tree structures on different, but current data associators require definition with mixins every time you want to use them. I think...

enhancement
good first issue

A few times a use case has come up to use Stone Soup metrics to investigate existing track and truth data sets, but this can be difficult if the track/truth...

enhancement

Should enable, via command line, to run end to end tracker from a configuration file.

Following #367, move functions in `__init__.py` into logical sub-modules, where enough similar functions can be grouped.

This allows each detection to belong to a different zone. This also allows tracks across multiple UTM zones to be compared to detections, and will then also allow tracks to...

enhancement

Many use cases requiring using custom data, beyond just CSV, so an example of this would be very useful.

documentation

`YAMLWriter` documentation is missing, making use unclear without looking at the code.