euclidean_probability

Find Probability of Match Based on Similarity

Empowers users to fuzzily-merge data frames with millions or tens of millions of rows in minutes with low memory usage. The package uses the locality sensitive hashing algorithms developed by Datar, Immorlica, Indyk and Mirrokni (2004) <doi:10.1145/997817.997857>, and Broder (1998) <doi:10.1109/SEQUEN.1997.666900> to avoid having to compare every pair of records in each dataset, resulting in fuzzy-merges that finish in linear time.

Beniamino Green

zoomerjoin

Superlatively Fast Fuzzy Joins

Etienne Bacher

The authors of the dependency Rust crates 

euclidean_probability function

<dl><dt>distance</dt>
<dd>the euclidian distance between the two vectors you want to
compare.</dd>
<dt>n_bands</dt>
<dd>The number of LSH bands used in hashing.</dd>
<dt>band_width</dt>
<dd>The number of hashes in each band.</dd>
<dt>r</dt>
<dd>the "r" hyperparameter used to govern the sensitivity of the hash.</dd></dl>

Arguments

Find Probability of Match Based on Similarity — euclidean_probability

<dl>

<dt>distance</dt>
<dd>the euclidian distance between the two vectors you want to
compare.</dd>


<dt>n_bands</dt>
<dd>The number of LSH bands used in hashing.</dd>


<dt>band_width</dt>
<dd>The number of hashes in each band.</dd>


<dt>r</dt>
<dd>the "r" hyperparameter used to govern the sensitivity of the hash.</dd>

</dl>

euclidean_probability: Find Probability of Match Based on Similarity

Description

Usage

Value

Arguments