Skip to main content

Most used methods for recommendation engine

Project description

About this package

Collaborative Filtering is a method that is often used as a recommendation engine. Many industries have used this algorithm to recommend their products, this library can use cosine similarity or even centered cosine similarity.

Depedencies

  • Python >= 3
  • numpy

How to use

  • cos_similarity(var1, var2, centered)

Parameters

var1 : iterable, array, np.array

The single arrray value you want to compare

var2 : iterable, array, np.array

The multi arrray values, you can convert from dataframe or table on your database

centered : bool

By default it will give true, if true you will use centered cosine similarity

Example case

  • We'll try to figuring out this single matrix value, on the multi array

[4, 0, 0, 5, 1, 0, 0]

  • The example dataset that we had

[[5, 5, 4, 0, 0, 0, 0],
[0, 0, 0, 2, 4, 5, 0],
[0, 3, 0, 0, 0, 0, 3]]

Example of code

from rengine.method import CollaborativeFiltering
import numpy as np
clf =  CollaborativeFiltering()
print(clf.cos_similarity(np.array([5,  4,  0,  0]),  [[1,  0,  3,  2],  [2,  1,  1,  0],  [4,  5,  0,  1]]))

Example of output

[-0.5, 0.866, -0.24] #row 2 in multi array data has more similarity

Another of output

[0.092, -0.559, nan] #nan means there's a number divided by zero

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

rengine-0.1.5.tar.gz (3.8 kB view details)

Uploaded Source

File details

Details for the file rengine-0.1.5.tar.gz.

File metadata

  • Download URL: rengine-0.1.5.tar.gz
  • Upload date:
  • Size: 3.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.4.1 importlib_metadata/3.10.0 pkginfo/1.7.0 requests/2.25.1 requests-toolbelt/0.9.1 tqdm/4.59.0 CPython/3.9.1

File hashes

Hashes for rengine-0.1.5.tar.gz
Algorithm Hash digest
SHA256 c2eb09938cdf9379828c6867bd27f841e5d310e0280aba591cd1ea98e5065073
MD5 ad6d3409e32a63856884aedbf2e3e564
BLAKE2b-256 d6855f985f785cb7a54f91939bbb25e86c28623e183d455100e6b4e23d00f951

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page