numenta / cortipy Goto Github PK

The REST API sometimes returns very sparse fingerprints -- for very uncommon terms, but also some more frequent terms, e.g. "after". Theses sparsities are ~0.1%, where the target for the API is 1.0%. Some viable solutions to increase the sparsity are:

1. randomly generate a fingerprint, as we do when the API gives a null response
2. preserving the few returned bits, fill out the sparsity w/ random generation
3. preserving the few returned bits, fill out the sparsity by returning the bits in closest (spatial) proximity to the returned bits

@subutai @scottpurdy what do you think?

Note: the previous HTTP API client, pycept, would scrap the returned bits and generate a new, random SDR at the target sparsity.

Investigate the new Cortical.io classification endpoint in their python SDK, and see if it is useful to add to cortipy.

Add simple examples to the readme.

The Cortical.io API now includes a logger at level INFO, printing out [INFO] Starting new HTTP connection (1): api.cortical.io for every query. We would like to suppress this info from printing to the console.

Add text classification example/tutorial, specifically using the NuPIC kNN classifier.

• also gives NuPIC a much-needed classification demo
• iPython walkthrough would be good

Currently the client raises an Exception for any response code that is not 200. This should be made more specific.

Caching of data returned from queries to Cortical.io API should be specific to the retina.