implicit

Fast Python Collaborative Filtering for Implicit Feedback Datasets

https://github.com/benfred/implicit

Science Score: 33.0%

This score indicates how likely this project is to be science-related based on various indicators:

○
CITATION.cff file
✓
codemeta.json file
Found codemeta.json file
○
.zenodo.json file
○
DOI references
✓
Academic publication links
Links to: arxiv.org
✓
Committers with academic emails
4 of 35 committers (11.4%) from academic institutions
○
Institutional organization owner
○
JOSS paper metadata
○
Scientific vocabulary similarity
Low similarity (15.6%) to scientific vocabulary

Keywords

collaborative-filtering machine-learning matrix-factorization recommendation recommendation-system recommender-system

Keywords from Contributors

optimizing-compiler tensors jax

Last synced: 6 months ago · JSON representation

Repository

Fast Python Collaborative Filtering for Implicit Feedback Datasets

Basic Info

Host: GitHub
Owner: benfred
License: mit
Language: Python
Default Branch: main
Homepage: https://benfred.github.io/implicit/
Size: 7.51 MB

Statistics

Stars: 3,702
Watchers: 76
Forks: 622
Open Issues: 103
Releases: 7

Topics

collaborative-filtering machine-learning matrix-factorization recommendation recommendation-system recommender-system

Created almost 10 years ago · Last pushed over 1 year ago

Metadata Files

Readme Changelog License

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

```python 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(useritemdata)

recommend items for a user

recommendations = model.recommend(userid, useritemdata[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: * H&M Personalized Fashion Recommendations Kaggle Competition * Yandex Cup 2022: Like Prediction * Recommending GitHub Repositories with Google BigQuery and the implicit library * Intro to Implicit Matrix Factorization: Classic ALS with Sketchfab Models * A Gentle Introduction to Recommender Systems with Implicit Feedback.

Requirements

This library requires SciPy version 0.16 or later and Python version 3.6 or later.

GPU Support requires at least version 11 of the NVidia CUDA Toolkit.

This library is tested with Python 3.7, 3.8, 3.9, 3.10 and 3.11 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 OPENBLASNUMTHREADS=1'. This disables its internal multithreading ability, which leads to substantial speedups for this package. Likewise for Intel MKL, setting 'export MKLNUMTHREADS=1' should also be set.

Released under the MIT License

Owner

Name: Ben Frederickson
Login: benfred
Kind: user
Location: Vancouver, BC, Canada
Company: @nvidia

Website: https://www.benfrederickson.com
Repositories: 39
Profile: https://github.com/benfred

Working on open source data science tools @nvidia

GitHub Events

Total

Issues event: 16
Watch event: 170
Issue comment event: 38
Pull request event: 2
Pull request review event: 4
Fork event: 15

Last Year

Issues event: 16
Watch event: 170
Issue comment event: 38
Pull request event: 2
Pull request review event: 4
Fork event: 15

Committers

Last synced: 9 months ago

All Time

Total Commits: 359
Total Committers: 35
Avg Commits per committer: 10.257
Development Distribution Score (DDS): 0.237

Past Year

Commits: 0
Committers: 0
Avg Commits per committer: 0.0
Development Distribution Score (DDS): 0.0

Top Committers

Name	Email	Commits
Ben Frederickson	g**b@b**m	274
Hyunsung Lee	i**a@g**m	32
Eugene Scherba	e**a@g**m	6
ita	i**9@r**m	5
Tomasz Cheda	c**z@g**m	3
yanz	d**7@g**m	3
Andrew Burkard	a**d@t**m	2
Christian Hotz-Behofsits	c**s@g**m	2
Juarez Bochi	j**i@g**m	2
Mark Douthwaite	m**7@y**k	2
Neal Fultz	n**z@g**m	2
Tim Gates	t**s@i**m	2
제프리	j**y@b**m	2
titipata	t**a@u**u	1
iggy.lee	i**e@k**m	1
Egor Malykh	e**h@o**v	1
Dmytro Petruk	p**k@e**h	1
mrticker	1****r	1
ds2268	d**8@s**i	1
Yurij Mikhalevich	0@3****t	1
Wei Zheng	i**1@g**m	1
Tych0n	N**v@g**m	1
Taylor G Smith	t**1@g**m	1
Martin Thoma	i**o@m**e	1
Kwon Soonmok	t**0@g**m	1
Jonas Johansson	j**s@g**m	1
James McNeilis	j**1@d**k	1
Iuri Queiroz	i**z@i**r	1
Dmytro Petruk	b**5@g**m	1
Daniel Julius Lasiman	d**n@g**m	1
and 5 more...

Committer Domains (Top 20 + Academic)

ifood.com.br: 1 dcs.bbk.ac.uk: 1 martin-thoma.de: 1 39.yt: 1 student.uni-lj.si: 1 ebu.ch: 1 okko.tv: 1 kakaocorp.com: 1 u.northwestern.edu: 1 buzzni.com: 1 iress.com: 1 york.ac.uk: 1 triplebyte.com: 1 ryencatchers.com: 1 benfrederickson.com: 1

Issues and Pull Requests

Last synced: 6 months ago

All Time

Total issues: 133
Total pull requests: 62
Average time to close issues: 5 months
Average time to close pull requests: 21 days
Total issue authors: 119
Total pull request authors: 15
Average comments per issue: 3.79
Average comments per pull request: 0.37
Merged pull requests: 47
Bot issues: 0
Bot pull requests: 1

Past Year

Issues: 14
Pull requests: 6
Average time to close issues: 3 days
Average time to close pull requests: N/A
Issue authors: 12
Pull request authors: 3
Average comments per issue: 0.93
Average comments per pull request: 0.17
Merged pull requests: 0
Bot issues: 0
Bot pull requests: 0

View more stats

Top Authors

Issue Authors

fkurushin (3)
Blo0dR0gue (3)
hemmat1991 (2)
Selva163 (2)
franzoni315 (2)
zakirullin (2)
mrticker (2)
LeGabriel (2)
josh-ashkinaze (2)
benfred (2)
levrone1987 (2)
lyanv (1)
jnhyeon (1)
jperiodlangley (1)
portkeys (1)

Pull Request Authors

benfred (40)
jmorlock (4)
bos1988 (3)
ita9naiwa (3)
mgorny (2)
characat0 (2)
Soonmok (1)
apat1n (1)
timgates42 (1)
jbochi (1)
atakanfilgoz (1)
dependabot[bot] (1)
mrticker (1)
chrisjkuch (1)
lithammer (1)

Top Labels

Issue Labels

enhancement (1)

Pull Request Labels

bug (12) enhancement (12) chore (7) dependencies (6) documentation (5) performance (1) ci (1)

Packages

Total packages: 5
Total downloads:
- pypi 405,678 last-month
Total docker downloads: 713

Total dependent packages: 12
(may contain duplicates)
Total dependent repositories: 196
(may contain duplicates)
Total versions: 87
Total maintainers: 4

pypi.org: implicit

Collaborative Filtering for Implicit Feedback Datasets

Homepage: http://github.com/benfred/implicit/
Documentation: https://implicit.readthedocs.io/
License: MIT
Latest release: 0.7.2
published over 2 years ago

Versions: 46
Dependent Packages: 12
Dependent Repositories: 195
Downloads: 405,640 Last month
Docker Downloads: 713

Rankings

Dependent packages count: 0.7%

Downloads: 0.9%

Dependent repos count: 1.1%

Average: 1.3%

Stargazers count: 1.3%

Docker downloads count: 1.7%

Forks count: 2.1%

Maintainers (2)

benfred ita9naiwa

Last synced: 6 months ago

proxy.golang.org: github.com/benfred/implicit

Documentation: https://pkg.go.dev/github.com/benfred/implicit#section-documentation
License: mit
Latest release: v0.7.2
published over 2 years ago

Versions: 21
Dependent Packages: 0
Dependent Repositories: 0

Rankings

Dependent packages count: 7.0%

Average: 8.2%

Dependent repos count: 9.3%

Last synced: 6 months ago

pypi.org: nautik-als

Collaborative Filtering for Implicit Datasets

Homepage: http://github.com/benfred/implicit/
Documentation: https://nautik-als.readthedocs.io/
License: MIT
Latest release: 0.0.5
published almost 6 years ago

Versions: 5
Dependent Packages: 0
Dependent Repositories: 1
Downloads: 10 Last month

Rankings

Stargazers count: 1.3%

Forks count: 2.1%

Dependent packages count: 7.3%

Dependent repos count: 22.1%

Average: 22.6%

Downloads: 80.1%

Maintainers (1)

xibin_zhu

Last synced: 6 months ago

pypi.org: wbimplicit

Collaborative Filtering for Implicit Feedback Datasets

Homepage: http://github.com/benfred/implicit/
Documentation: https://wbimplicit.readthedocs.io/
License: MIT
Latest release: 0.6.2.1
published over 2 years ago

Versions: 1
Dependent Packages: 0
Dependent Repositories: 0
Downloads: 28 Last month

Rankings

Stargazers count: 1.3%

Forks count: 2.1%

Dependent packages count: 7.5%

Average: 25.3%

Downloads: 45.8%

Dependent repos count: 69.6%

Maintainers (1)

gallir

Last synced: 6 months ago

conda-forge.org: implicit

This project provides fast Python implementations of the algorithms described in the paper Collaborative Filtering for Implicit Feedback Datasets and in Applications of the Conjugate Gradient Method for Implicit Feedback Collaborative Filtering

Homepage: http://github.com/benfred/implicit/
License: MIT
Latest release: 0.5.2
published about 4 years ago

Versions: 14
Dependent Packages: 0
Dependent Repositories: 0

Rankings

Forks count: 6.7%

Stargazers count: 7.2%

Average: 26.0%

Dependent repos count: 38.6%

Dependent packages count: 51.5%

Last synced: 6 months ago

Dependencies

.github/workflows/build.yml actions

actions/checkout v3 composite
actions/download-artifact v3 composite
actions/setup-python v4 composite
actions/setup-python v2 composite
actions/upload-artifact v3 composite
docker/setup-qemu-action v2 composite
fnkr/github-action-ghr v1.3 composite
pypa/cibuildwheel v2.11.3 composite

.github/workflows/release-drafter.yml actions

release-drafter/release-drafter v5 composite

requirements-dev.txt pypi

black ==22.3.0 development
codespell * development
flake8 * development
isort * development
pylint * development
pytest * development
scikit-build >=0.13.1 development

requirements.txt pypi

Cython >=0.24.0
scipy >=0.16.0
tqdm >=4.27.0

setup.py pypi

numpy *

pyproject.toml pypi

implicit

Science Score: 33.0%

Keywords

Keywords from Contributors

Repository

Basic Info

Statistics

Topics

Metadata Files

README.md

Implicit

Installation

CPU only package

CPU+GPU package

Basic Usage

initialize a model

train the model on a sparse matrix of user/item/confidence weights

recommend items for a user

find related items

Articles about Implicit

Requirements

Benchmarks

Optimal Configuration

Owner

GitHub Events

Total

Last Year

Committers

All Time

Past Year

Top Committers

Committer Domains (Top 20 + Academic)

Issues and Pull Requests

All Time

Past Year

Top Authors

Issue Authors

Pull Request Authors

Top Labels

Issue Labels

Pull Request Labels

Packages

pypi.org: implicit

Rankings

Maintainers (2)

proxy.golang.org: github.com/benfred/implicit

Rankings

pypi.org: nautik-als

Rankings

Maintainers (1)

pypi.org: wbimplicit

Rankings

Maintainers (1)

conda-forge.org: implicit

Rankings

Dependencies