more robust divergence caching

djsutherland / py-sdm

Python implementation of nonparametric nearest-neighbor-based estimators for divergences between distributions.

http://cs.cmu.edu/~dsutherl/sdm/

BSD 3-Clause "New" or "Revised" License

48 stars 8 forks source link

Open djsutherland opened 11 years ago

djsutherland commented 11 years ago

Supporting:

loading (and saving?) partial results based on categories/names
actually working with multiple values of k, different specs (unlike sdm.sdm.get_divs_cache)

Also: maybe the default names should include the number of points as well, or even a hash or something, to identify them more robustly?

djsutherland commented 11 years ago

Hashing features for the contents in cache files is a good idea. Could use cityhash (python bindings: Amper/cityhash) or just md5 or whatever.