by An Uncommon Lab


Deploy filters generated by the *kf engine along with any necessary utilities so that a user can run the filters on a machine that does not have *kf installed.

skfdeploy(zip_name, filter_names);
skfdeploy(zip_name, filter_names, other_files);

For instance, suppose you generated two filters using the *kf engine, one called 'filter_a' and the other called 'filter_b'. These can be zipped up with:

skfdeploy('my_filters.zip', {'filter_a'; 'filter_b'});

There's no need to identify the specific elements of the filter that are generated, such as calling out the 'filter_a_step.m' file; skfdeploy will find those automatically.

The generated filters must be on the path in order skfdeploy to find them.

This utility does not include user-generated files, such as the user's propagation function; those are usually distributed by some other means, such as a version control system. However, if they are required, they may be included as additional files, as below.

For convenience, any other files that should be included with the zip can be passed as a cell array of strings. For instance, if we wanted to include the user's propagation and observation functions, my_prop.m and my_obs.m, in the above zip, we would use this command:

skfdeploy('my_filters.zip', ...
          {'filter_a'; 'filter_b'}, ...
          {'my_prop.m', 'my_obs.m'});

The function will return a list of file names appearing in the zip file.

See Also

skfengine, skfgen

Table of Contents

  1. See Also