Collaborative Filtering for Implicit Feedback Datasets

These details have not been verified by PyPI

Project description

Implicit

Fast Python Collaborative Filtering for Implicit Datasets.

This project provides fast Python implementations of several different popular recommendation algorithms for implicit feedback datasets:

Alternating Least Squares as described in the papers Collaborative Filtering for Implicit Feedback Datasets and Applications of the Conjugate Gradient Method for Implicit Feedback Collaborative Filtering.
Bayesian Personalized Ranking.
Logistic Matrix Factorization
Item-Item Nearest Neighbour models using Cosine, TFIDF or BM25 as a distance metric.

All models have multi-threaded training routines, using Cython and OpenMP to fit the models in parallel among all available CPU cores. In addition, the ALS and BPR models both have custom CUDA kernels - enabling fitting on compatible GPU's. Approximate nearest neighbours libraries such as Annoy, NMSLIB and Faiss can also be used by Implicit to speed up making recommendations.

Installation

Implicit can be installed from pypi with:

pip install implicit

Installing with pip will use prebuilt binary wheels on x86_64 Linux, Windows and OSX. These wheels include GPU support on Linux.

Implicit can also be installed with conda:

# CPU only package
conda install -c conda-forge implicit

# CPU+GPU package
conda install -c conda-forge implicit implicit-proc=*=gpu

Basic Usage

import implicit

# initialize a model
model = implicit.als.AlternatingLeastSquares(factors=50)

# train the model on a sparse matrix of user/item/confidence weights
model.fit(user_item_data)

# recommend items for a user
recommendations = model.recommend(userid, user_item_data[userid])

# find related items
related = model.similar_items(itemid)

The examples folder has a program showing how to use this to compute similar artists on the last.fm dataset.

For more information see the documentation.

Articles about Implicit

These blog posts describe the algorithms that power this library:

There are also several other articles about using Implicit to build recommendation systems:

Requirements

This library requires SciPy version 0.16 or later and Python version 3.9 or later.

GPU Support requires version 13 of the NVidia CUDA Toolkit, and requires that RMM be installed pip install rmm-cu13

This library is tested with Python 3.9 to 3.14 on Ubuntu, OSX and Windows.

Benchmarks

Simple benchmarks comparing the ALS fitting time versus Spark can be found here.

Optimal Configuration

I'd recommend configuring SciPy to use Intel's MKL matrix libraries. One easy way of doing this is by installing the Anaconda Python distribution.

For systems using OpenBLAS, I highly recommend setting 'export OPENBLAS_NUM_THREADS=1'. This disables its internal multithreading ability, which leads to substantial speedups for this package. Likewise for Intel MKL, setting 'export MKL_NUM_THREADS=1' should also be set.

Released under the MIT License

Project details

These details have not been verified by PyPI

Release history Release notifications | RSS feed

This version

0.7.3

May 8, 2026

0.7.2

Sep 29, 2023

0.7.1

Aug 25, 2023

0.7.0

Jun 13, 2023

0.6.2

Dec 11, 2022

0.6.1

Aug 31, 2022

0.6.0

Jul 11, 2022

0.5.2

Jan 29, 2022

0.5.1

Jan 28, 2022

0.5.0

Jan 28, 2022

0.4.8

Aug 29, 2021

0.4.7

Aug 29, 2021

0.4.6

Aug 29, 2021

0.4.5

Aug 29, 2021

0.4.4

Sep 15, 2020

0.4.3

Sep 15, 2020

0.4.2

Oct 23, 2019

0.4.0

Aug 1, 2019

0.3.9

Jul 13, 2019

0.3.8

Oct 11, 2018

0.3.7

Sep 22, 2018

0.3.6

May 24, 2018

0.3.5

May 8, 2018

0.3.4

Apr 7, 2018

0.3.3

Mar 15, 2018

0.3.2

Mar 14, 2018

0.3.1

Feb 10, 2018

0.3.0

Feb 8, 2018

0.2.13

Feb 8, 2018

0.2.12

Feb 8, 2018

0.2.11

Feb 3, 2018

0.2.10

Jan 31, 2018

0.2.9

Jan 29, 2018

0.2.8

Jan 29, 2018

0.2.7

Nov 1, 2017

0.2.6

Jun 25, 2017

0.2.5

May 15, 2017

0.2.4

May 14, 2017

0.2.3

Mar 26, 2017

0.2.2

Mar 2, 2017

0.2.1

Feb 25, 2017

0.1.7

Dec 13, 2016

0.1.5

Oct 20, 2016

0.1.4

Oct 6, 2016

0.1.3

Sep 21, 2016

0.1.1

Apr 24, 2016

0.1.0

Apr 17, 2016

Download files

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

Source Distribution

implicit-0.7.3.tar.gz (444.0 kB view details)

Uploaded May 14, 2026 Source

Built Distributions

If you're not sure about the file name format, learn more about wheel file names.

The dropdown lists show the available interpreters, ABIs, and platforms. Enable javascript to be able to filter the list of wheel files.

implicit-0.7.3-cp314-cp314-win_amd64.whl (692.7 kB view details)

Uploaded May 8, 2026 CPython 3.14Windows x86-64

implicit-0.7.3-cp314-cp314-manylinux2014_x86_64.whl (7.6 MB view details)

Uploaded May 8, 2026 CPython 3.14

implicit-0.7.3-cp314-cp314-manylinux2014_aarch64.whl (861.0 kB view details)

Uploaded May 8, 2026 CPython 3.14

implicit-0.7.3-cp314-cp314-macosx_11_0_arm64.whl (650.2 kB view details)

Uploaded May 8, 2026 CPython 3.14macOS 11.0+ ARM64

implicit-0.7.3-cp314-cp314-macosx_10_15_x86_64.whl (679.6 kB view details)

Uploaded May 8, 2026 CPython 3.14macOS 10.15+ x86-64

implicit-0.7.3-cp314-cp314-macosx_10_15_universal2.whl (1.3 MB view details)

Uploaded May 8, 2026 CPython 3.14macOS 10.15+ universal2 (ARM64, x86-64)

implicit-0.7.3-cp313-cp313-win_amd64.whl (675.3 kB view details)

Uploaded May 8, 2026 CPython 3.13Windows x86-64

implicit-0.7.3-cp313-cp313-manylinux2014_x86_64.whl (7.6 MB view details)

Uploaded May 8, 2026 CPython 3.13

implicit-0.7.3-cp313-cp313-manylinux2014_aarch64.whl (851.8 kB view details)

Uploaded May 8, 2026 CPython 3.13

implicit-0.7.3-cp313-cp313-macosx_11_0_arm64.whl (644.1 kB view details)

Uploaded May 8, 2026 CPython 3.13macOS 11.0+ ARM64

implicit-0.7.3-cp313-cp313-macosx_10_13_x86_64.whl (675.9 kB view details)

Uploaded May 8, 2026 CPython 3.13macOS 10.13+ x86-64

implicit-0.7.3-cp313-cp313-macosx_10_13_universal2.whl (1.3 MB view details)

Uploaded May 8, 2026 CPython 3.13macOS 10.13+ universal2 (ARM64, x86-64)

implicit-0.7.3-cp312-cp312-win_amd64.whl (676.7 kB view details)

Uploaded May 8, 2026 CPython 3.12Windows x86-64

implicit-0.7.3-cp312-cp312-manylinux2014_x86_64.whl (7.6 MB view details)

Uploaded May 8, 2026 CPython 3.12

implicit-0.7.3-cp312-cp312-manylinux2014_aarch64.whl (843.6 kB view details)

Uploaded May 8, 2026 CPython 3.12

implicit-0.7.3-cp312-cp312-macosx_11_0_arm64.whl (647.8 kB view details)

Uploaded May 8, 2026 CPython 3.12macOS 11.0+ ARM64

implicit-0.7.3-cp312-cp312-macosx_10_13_x86_64.whl (679.7 kB view details)

Uploaded May 8, 2026 CPython 3.12macOS 10.13+ x86-64

implicit-0.7.3-cp312-cp312-macosx_10_13_universal2.whl (1.3 MB view details)

Uploaded May 8, 2026 CPython 3.12macOS 10.13+ universal2 (ARM64, x86-64)

implicit-0.7.3-cp311-cp311-win_amd64.whl (692.9 kB view details)

Uploaded May 8, 2026 CPython 3.11Windows x86-64

implicit-0.7.3-cp311-cp311-manylinux2014_x86_64.whl (7.7 MB view details)

Uploaded May 8, 2026 CPython 3.11

implicit-0.7.3-cp311-cp311-manylinux2014_aarch64.whl (874.8 kB view details)

Uploaded May 8, 2026 CPython 3.11

implicit-0.7.3-cp311-cp311-macosx_11_0_arm64.whl (641.8 kB view details)

Uploaded May 8, 2026 CPython 3.11macOS 11.0+ ARM64

implicit-0.7.3-cp311-cp311-macosx_10_9_x86_64.whl (671.5 kB view details)

Uploaded May 8, 2026 CPython 3.11macOS 10.9+ x86-64

implicit-0.7.3-cp311-cp311-macosx_10_9_universal2.whl (1.2 MB view details)

Uploaded May 8, 2026 CPython 3.11macOS 10.9+ universal2 (ARM64, x86-64)

implicit-0.7.3-cp310-cp310-win_amd64.whl (693.1 kB view details)

Uploaded May 8, 2026 CPython 3.10Windows x86-64

implicit-0.7.3-cp310-cp310-manylinux2014_x86_64.whl (7.7 MB view details)

Uploaded May 8, 2026 CPython 3.10

implicit-0.7.3-cp310-cp310-manylinux2014_aarch64.whl (874.9 kB view details)

Uploaded May 8, 2026 CPython 3.10

implicit-0.7.3-cp310-cp310-macosx_11_0_arm64.whl (644.0 kB view details)

Uploaded May 8, 2026 CPython 3.10macOS 11.0+ ARM64

implicit-0.7.3-cp310-cp310-macosx_10_9_x86_64.whl (676.6 kB view details)

Uploaded May 8, 2026 CPython 3.10macOS 10.9+ x86-64

implicit-0.7.3-cp310-cp310-macosx_10_9_universal2.whl (1.3 MB view details)

Uploaded May 8, 2026 CPython 3.10macOS 10.9+ universal2 (ARM64, x86-64)

implicit-0.7.3-cp39-cp39-win_amd64.whl (695.0 kB view details)

Uploaded May 8, 2026 CPython 3.9Windows x86-64

implicit-0.7.3-cp39-cp39-manylinux2014_x86_64.whl (7.7 MB view details)

Uploaded May 8, 2026 CPython 3.9

implicit-0.7.3-cp39-cp39-manylinux2014_aarch64.whl (876.8 kB view details)

Uploaded May 8, 2026 CPython 3.9

implicit-0.7.3-cp39-cp39-macosx_11_0_arm64.whl (646.0 kB view details)

Uploaded May 8, 2026 CPython 3.9macOS 11.0+ ARM64

implicit-0.7.3-cp39-cp39-macosx_10_9_x86_64.whl (678.7 kB view details)

Uploaded May 8, 2026 CPython 3.9macOS 10.9+ x86-64

implicit-0.7.3-cp39-cp39-macosx_10_9_universal2.whl (1.3 MB view details)

Uploaded May 8, 2026 CPython 3.9macOS 10.9+ universal2 (ARM64, x86-64)

File details

Details for the file implicit-0.7.3.tar.gz.

File metadata

Download URL: implicit-0.7.3.tar.gz
Upload date: May 14, 2026
Size: 444.0 kB
Tags: Source
Uploaded using Trusted Publishing? No
Uploaded via: twine/6.2.0 CPython/3.14.4

File hashes

Hashes for implicit-0.7.3.tar.gz
Algorithm	Hash digest
SHA256	`9571a52d70b3667e99a6e6aae6681526dd3286c765cf0ea10e1073d068a98057`
MD5	`f8e0f6e4bc35010a1b3e7d5c5e6fd392`
BLAKE2b-256	`3224bdaa4be7fcfb568d63e0ddff948c486515c6b728055ee060f3f131803476`

Algorithm	Hash digest
SHA256	`bf7424055ae22289e6d519b7e8bc31e7af8ea1e07a0ca6fd85c4d2f4822f3ae3`
MD5	`3e8a0d4e84a964060db913da0b6ec1d6`
BLAKE2b-256	`5c2c4e0fb43210462b5c06e50f61780fa88ffd62bbc4c34516219d8c75ed0a54`

Algorithm	Hash digest
SHA256	`ec1f2c451685898de0379e12da29e6a50391164f4b21760cb4f06a35ff1a8e8f`
MD5	`980ecba5f86f20fbdd4815ff08f5edb5`
BLAKE2b-256	`4d6ac4fda48bfe88ba0ad6afc19cd100bff37e692ffc13be60aefb0aaf12a135`

Algorithm	Hash digest
SHA256	`7518514aaeb3348103a8bb31e15e45d1af1c4564403dd9dbe7d09c5e85ad8b83`
MD5	`25a83e6ab0f39bdcc237682083aa9354`
BLAKE2b-256	`94d950d64ed847a75afa8d58d754b07311cfe87dbdf1b2f5eb3e7664a69e162f`

Algorithm	Hash digest
SHA256	`53f6607bc2ec2c5f4b032e7a904a8cec7c6aa552bd844a69ad46ee65225cb1e0`
MD5	`9ec3c99d88a8659dfe2f8777b7b143d4`
BLAKE2b-256	`b8818fff25d12d58933edeb480a8f7ec8c5e4f971b0313c00b9e996ea1b70f33`

Algorithm	Hash digest
SHA256	`7299d2d127ba86199c5794d17db923f45d9ffe1b4eb5cdd3ba5aa64600dfbe2d`
MD5	`0ecc1761185f52332a95c95f2ce09fc6`
BLAKE2b-256	`e2d68784c71e7de78df3e5efbf7a0429e9b0cf51f5dedceb19ebe9e2351a9c9a`

Algorithm	Hash digest
SHA256	`30b46876e27add6d65125fd9dc0b9af4a1b5667123cf76aefbd38bcdd23e69b4`
MD5	`3eb4b105ccd2b68580d7bf9dafba4ba1`
BLAKE2b-256	`8a51508fd3b18f10499bac710fdcdf2832fe4ca9d9c0e98702dab3d4761a7270`

Algorithm	Hash digest
SHA256	`395926a937cc75e89eecd5b9042df18b6b2dca826bb7e11b2abdff7584a2ea4e`
MD5	`9385978cc770ce5b0faf626f5b852e7e`
BLAKE2b-256	`95f7d1172da1580dd5a51fd15b71146d12bc1b93c6312e598189360c4ca8ba8b`

Algorithm	Hash digest
SHA256	`71573a1083b33878032c44e4caa04c6853b0f1c959bf5460f4fb07e396267fc4`
MD5	`165f7189c693e4e31901ccb4efddfe0f`
BLAKE2b-256	`f15f9e4a773cfa80409af06b89c98b642595a280a7648b70eed5aefd049b1d77`

Algorithm	Hash digest
SHA256	`6a2caddac3ca46840bb8db5f8a7d386b00c9c708095e8df86919c737c3ceca8e`
MD5	`930d1cef2c59995cea2ed2476ced8e97`
BLAKE2b-256	`e0616aa8760fa328f9ffcbf1104475819c0cfbe5f76ea260c8ee5b9dfbbeaeaa`

Algorithm	Hash digest
SHA256	`00a7791b7b0a6a0802c95dc88a432e5af6beb19cd4a948196df98fb42a2a7d96`
MD5	`61b95a700b1a2a55494a05ea54d2e39a`
BLAKE2b-256	`f08a093ed80fff27ad9c188daeec68bbea507345c5fcfab58442e0e42430a653`

Algorithm	Hash digest
SHA256	`36bb093bad91479e9226846de59430848165c6e423c8959b191e009f04bd8619`
MD5	`1135c32fef413268ff086de86e54abca`
BLAKE2b-256	`502ee4c3a7ba2c144c9b237d15f9eb77b4df1011f287f0d6cbd41b75c5cc9e05`

Algorithm	Hash digest
SHA256	`193ac8dc10614fc0db153a0bf3a7f53f4cbad094c1b7938c7f22c796e541a3bb`
MD5	`ff3f27a140c04a36a72aa8a55aa11ce7`
BLAKE2b-256	`0eba98265ef5379302f5d694585bdf1cc7b0ab90f6e54d06466d2a0e5019bd56`

Algorithm	Hash digest
SHA256	`c3e10e634a1b24bb1e8a036529940b8884c33a2dbb9fb19194827c0b68802edc`
MD5	`39e43fe769ed3fc75f5ebb55a975ccb9`
BLAKE2b-256	`9544d8de15893da635602dd2880855494c775d14a0380641d7082fd8efa1a02f`

Algorithm	Hash digest
SHA256	`0de3734a74db5248b474ebf2c2e9dcf9ab16de137a009436f6ae8f8880486ac2`
MD5	`de73eccb99f7bf3ce4a7dcad6ef7490b`
BLAKE2b-256	`fd4a615b55e3303fee9d1f54a0e01ea399712f77774fccbe763d125a51d9b978`

Algorithm	Hash digest
SHA256	`ab5825723918b790cab45c2ec64669ac03b1b1774e9ce62957b67ad37fdf2d45`
MD5	`99303a6ffd7c18717ea0e7291decc815`
BLAKE2b-256	`654be4c93654e6c5303a544f4505fb25a962213d603886ce71558026ce698cbd`

Algorithm	Hash digest
SHA256	`4483a3746d97f077c61d87a23bc7b6948ae75957def1cb4828d4a96ec5b91e75`
MD5	`fe3e0a50b30fd14fc93e9257817f5e50`
BLAKE2b-256	`7c25b3db94464aa01bee2a3717217d425d3ba8f1e18689ed01ac000c79d88cb9`

Algorithm	Hash digest
SHA256	`a04066fa5798bb094533678b45f1a3298cd44acb8c42f516fe46893be1989227`
MD5	`f1847641eb4ae6bfe0bf88b938fd1b0f`
BLAKE2b-256	`8d43e398cf7b789a540f9143296c03204959c1d9e1b1bdb9bd176d6b5e55afd8`

Algorithm	Hash digest
SHA256	`4970db19607f6f33cc4f2567233e03a4902191b6ebed7f264f8810d730dadb23`
MD5	`405f1011d41ddfc0d68d31fc7c922ef8`
BLAKE2b-256	`b3dcbc39f9df23682e31ce5b09d00f5c92591a8d750f84b3d4751d07f6ea61a7`

Algorithm	Hash digest
SHA256	`02cf46507f62514415374511cc79b2cba8021d9d432abfca4407276d4d41bc4d`
MD5	`ac2c321ae117d36c7750ce05f5a6b669`
BLAKE2b-256	`8db0a0cb666d80520bf6a5be4370cd7f572b7a4057fc7a3779249f10d1dec840`

Algorithm	Hash digest
SHA256	`bad1d7263ce295674562cf74482a83ae6847579579f8e2f5e70da840386c2c14`
MD5	`923ae8da571897f4bf1c9087d355f3ff`
BLAKE2b-256	`316e9d0e5a300bd20847202e071030147ec9db0bcd9e3e2d2ba44ba0153836b8`

implicit 0.7.3

Navigation

Verified details

Maintainers

Unverified details

Meta

Classifiers

Project description

Implicit

Installation

Basic Usage

Articles about Implicit

Requirements

Benchmarks

Optimal Configuration

Project details

Verified details

Maintainers

Unverified details

Meta

Classifiers

Release history Release notifications | RSS feed

Download files

Source Distribution

Built Distributions

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

File metadata

File hashes

File details

Algorithm	Hash digest
SHA256	`e4d56a2d9db92e9cf1d0c9af740581d40423c216b0e207d7d72d59a0ca5519c2`
MD5	`f5bce16bfeedabd602943e661a60991c`
BLAKE2b-256	`9ce2f4102085c133c0e824976a55d3b59f244029aad50e5305fa4c4d38654bc2`

Algorithm	Hash digest
SHA256	`77b9b93ea7840be5367c0b49ff90a371d5f5e86bc4743776ee491dfb62854c2b`
MD5	`ddb0e6798d2145b13e65ebf1196b7d03`
BLAKE2b-256	`e7300f855b8b74dfd441e07b26db2ec230d110fdb9276cef75b2911d29606d32`

Algorithm	Hash digest
SHA256	`7a6035c0fbe89eef9d1789367c8097eb3145af3cfffd0dc0e92b2570ab67a6fc`
MD5	`bf3a3e2c8d9a861a8e64a9c31c9f970e`
BLAKE2b-256	`067f16e15852849f99ccbaf9f4e0f8ccbab08f20ba37b09a60ad5b84e8d2e415`

Algorithm	Hash digest
SHA256	`a059c926f23813d288b0ad68e436c84ed24c7049a26dc1b4cced70c9b8d1cf40`
MD5	`77949d82123058942199d107fc5f3b87`
BLAKE2b-256	`81e45d033371a1aca288bba68ed65a74307cb1ec3f6bb03339b3df6b8c050d37`

Algorithm	Hash digest
SHA256	`0b4b35427fadf6c5183391ea913b5b2a08b886af90e40343743ca3490dbbed1f`
MD5	`3a1a38e60553095783ca42c735449ba0`
BLAKE2b-256	`b816f6fe2b0c669dcc77f12a497f271d2bc5afc9a9fedff16eee1f42dff0e3f2`

Algorithm	Hash digest
SHA256	`0217a6e8e2d2c217afde258e81d0dd53348e3a3ac36c298c151c7ebc722ae36d`
MD5	`6a91697ccacefdae85bf754b8cb6872a`
BLAKE2b-256	`d864ce02c12766b00a45b3b25fba29cd71b54b5cdc0a8adaa64b4b6c859ee0dc`

Algorithm	Hash digest
SHA256	`7d80d227d1f4d75e7cbbec7cde02560f732710537765fd3157afb14a87c46070`
MD5	`f0b909a2c4010968ad2e61318a7b4966`
BLAKE2b-256	`03b28e0fa67299c55801c9d7454f99d3971f5fc543e139c54f1a839d81cb2d34`

Algorithm	Hash digest
SHA256	`d479f91657af5dfade0c87a671303acfc954d8bb2aaf5622524783dff490596f`
MD5	`4a30047b8ccd70d5daf3e3bd07910f6e`
BLAKE2b-256	`b10d71fc8712d4a01867428cdd34527762e20bf28b831126ed9ef2633a416f15`

Algorithm	Hash digest
SHA256	`73346fe465ddbf9281627816300cc4249817bd5284ace023721717f22288656b`
MD5	`e0214a182819370fee472d0c66978673`
BLAKE2b-256	`f020b9926cfaa0a947da0d87bb60c5978d285686b60b583ffc5d8d43bf07aee3`

Algorithm	Hash digest
SHA256	`17a844e9783550d33148fda6b8e9067bb4448d4f2b154ec4ef97df24d16bccdd`
MD5	`9cdc285b6f788077016b277201db15f3`
BLAKE2b-256	`8a69cba8ac02a3f560dabca05e50455402ab761a30f1225ed4b85276487dcbd4`