Skip to main content

fast procedures for forking with hashes

Project description

In Deep Learning research, hashing, retrieval, and ranking tasks often require calculation of mAP of retrieval, which can be computationally expensive. Often, hashes are represented as ndarrays of floats, where the sign of the float number has the meaning of a bit. This happens because neural networks work with floats and that is precisely the output that they give. Computing ranking of documents in the DB based the query typically implemented in NumPy, which can be quite suboptimal.

This package provides several functions that allow fast hamming distance computation, ranking, and mAP computation.

All API has two backends: * NumPy implementation, which is simple, straightforward and as efficient as it can get in pure NumPy. Used as a reference. * C++ Python extension that implements the same API, on average 10x faster than NumPy implementation and significantly more memory efficient.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

hashranking-0.0.1.tar.gz (7.1 kB view hashes)

Uploaded source

Built Distributions

hashranking-0.0.1-cp36-cp36m-win_amd64.whl (295.0 kB view hashes)

Uploaded cp36

hashranking-0.0.1-cp27-cp27m-win_amd64.whl (137.2 kB view hashes)

Uploaded cp27

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Huawei Huawei PSF Sponsor Microsoft Microsoft PSF Sponsor NVIDIA NVIDIA PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page