A science toolkit for recommender systems

Project description

Scikit-Recommender

Scikit-Recommender is an open source library for researchers of recommender systems.

Highlighted Features

Various recommendation models
Parse arguments from command line and ini-style files
Diverse data preprocessing
Fast negative sampling
Fast model evaluation
Convenient record logging
Flexible batch data iterator

Installation

You have three ways to use Scikit-Recommender:

Install from PyPI
Install from Source
Run without Installation

Install from PyPI

Binary installers are available at the Python package index and you can install the package from pip.

pip install scikit-recommender

Install from Source

Installing from source requires Cython and the current code works well with the version 0.29.20.

To build scikit-recommender from source you need Cython:

pip install cython==0.29.20

Then, the scikit-recommender can be installed by executing:

git clone https://github.com/ZhongchuanSun/scikit-recommender.git
cd scikit-recommender
python setup.py install

Run without Installation

Alternatively, You can also run the sources without installation. Please compile the cython codes before running:

git clone https://github.com/ZhongchuanSun/scikit-recommender.git
cd scikit-recommender
python setup.py build_ext --inplace

Usage

After installing or compiling this package, now you can run the run_skrec.py:

python run_skrec.py

You can also find examples in tutorial.ipynb.

Models

MMRec	Implementation	Paper	Publication
MGCN	PyTorch	Penghang Yu, et al., Multi-View Graph Convolutional Network for Multimedia Recommendation	ACM MM 2023
BM3	PyTorch	Xin Zhou, et al., Bootstrap Latent Representations for Multi-modal Recommendation	WWW 2023
FREEDOM	PyTorch	Xin Zhou, et al., A Tale of Two Graphs: Freezing and Denoising Graph Structures for Multimodal Recommendation	ACM MM 2023
SLMRec	PyTorch	Zhulin Tao, et al., Self-supervised Learning for Multimedia Recommendation	TMM 2022
LATTICE	PyTorch	Jinghao Zhang, et al., Mining Latent Structures for Multimedia Recommendation	ACM MM 2021

Recommender	Implementation	Paper	Publication
SelfCF	PyTorch	Xin Zhou, et al., SelfCF: A Simple Framework for Self-supervised Collaborative Filtering	TORS 2023
LayerGCN	PyTorch	Xin Zhou, et al., Layer-refined Graph Convolutional Networks for Recommendation	ICDE 2023
DENS	PyTorch	Riwei Lai, et al., Disentangled Negative Sampling for Collaborative Filtering	WSDM 2023
LightGCL	PyTorch	Xuheng Cai, et al., LightGCL: Simple Yet Effective Graph Contrastive Learning for Recommendation	ICLR 2023
SGAT	TensorFlow (1.14)	Zhongchuan Sun, et al., Sequential Graph Collaborative Filtering	Information Sciences 2022
LightGCN	PyTorch	Xiangnan He et al., LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation.	SIGIR 2020
SRGNN	TensorFlow (1.14)	Shu Wu et al., Session-Based Recommendation with Graph Neural Networks.	AAAI 2019
HGN	PyTorch	Chen Ma et al., Hierarchical Gating Networks for Sequential Recommendation.	KDD 2019
BERT4Rec	TensorFlow (1.14)	Fei Sun et al., BERT4Rec: Sequential Recommendation with Bidirectional Encoder Representations from Transformer.	CIKM 2019
SASRec	TensorFlow (1.14)	Wangcheng Kang et al., Self-Attentive Sequential Recommendation.	ICDM 2018
GRU4RecPlus	TensorFlow (1.14)	Balázs Hidasi et al., Recurrent Neural Networks with Top-k Gains for Session-based Recommendations.	CIKM 2018
Caser	PyTorch	Jiaxi Tang et al., Personalized Top-N Sequential Recommendation via Convolutional Sequence Embedding.	WSDM 2018
MultiVAE	PyTorch	Dawen Liang, et al., Variational Autoencoders for Collaborative Filtering.	WWW 2018
TransRec	PyTorch	Ruining He et al., Translation-based Recommendation.	RecSys 2017
CML	TensorFlow (1.14)	Cheng-Kang Hsieh et al., Collaborative Metric Learning.	WWW 2017
CDAE	PyTorch	Yao Wu et al., Collaborative Denoising Auto-Encoders for Top-n Recommender Systems.	WSDM 2016
GRU4Rec	TensorFlow (1.14)	Balázs Hidasi et al., Session-based Recommendations with Recurrent Neural Networks.	ICLR 2016
AOBPR	C/Cython	Steffen Rendle et al., Improving Pairwise Learning for Item Recommendation from Implicit Feedback.	WSDM 2014
FPMC	PyTorch	Steffen Rendle et al., Factorizing Personalized Markov Chains for Next-Basket Recommendation.	WWW 2010
BPRMF	PyTorch	Steffen Rendle et al., BPR: Bayesian Personalized Ranking from Implicit Feedback.	UAI 2009
Pop	Python	Make recommendations based on item popularity.

Project details

Release history Release notifications | RSS feed

This version

0.1.1

Sep 29, 2023

0.1.0

Sep 29, 2023

0.0.5

Sep 26, 2023

0.0.4

Jun 12, 2023

0.0.3

May 30, 2023

0.0.2

May 29, 2023

0.0.1

Jul 13, 2021

Download files

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

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distributions

scikit_recommender-0.1.1-cp311-cp311-win_amd64.whl (360.1 kB view hashes)

Uploaded Sep 29, 2023 CPython 3.11 Windows x86-64

scikit_recommender-0.1.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (2.6 MB view hashes)

Uploaded Sep 29, 2023 CPython 3.11 manylinux: glibc 2.17+ x86-64

scikit_recommender-0.1.1-cp311-cp311-macosx_10_9_universal2.whl (637.4 kB view hashes)

Uploaded Sep 29, 2023 CPython 3.11 macOS 10.9+ universal2 (ARM64, x86-64)

scikit_recommender-0.1.1-cp310-cp310-win_amd64.whl (359.2 kB view hashes)

Uploaded Sep 29, 2023 CPython 3.10 Windows x86-64

scikit_recommender-0.1.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (2.5 MB view hashes)

Uploaded Sep 29, 2023 CPython 3.10 manylinux: glibc 2.17+ x86-64

scikit_recommender-0.1.1-cp310-cp310-macosx_11_0_x86_64.whl (395.8 kB view hashes)

Uploaded Sep 29, 2023 CPython 3.10 macOS 11.0+ x86-64

scikit_recommender-0.1.1-cp39-cp39-win_amd64.whl (391.3 kB view hashes)

Uploaded Sep 29, 2023 CPython 3.9 Windows x86-64

scikit_recommender-0.1.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (2.5 MB view hashes)

Uploaded Sep 29, 2023 CPython 3.9 manylinux: glibc 2.17+ x86-64

scikit_recommender-0.1.1-cp39-cp39-macosx_11_0_x86_64.whl (395.8 kB view hashes)

Uploaded Sep 29, 2023 CPython 3.9 macOS 11.0+ x86-64

scikit_recommender-0.1.1-cp38-cp38-win_amd64.whl (392.5 kB view hashes)

Uploaded Sep 29, 2023 CPython 3.8 Windows x86-64

scikit_recommender-0.1.1-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (2.5 MB view hashes)

Uploaded Sep 29, 2023 CPython 3.8 manylinux: glibc 2.17+ x86-64

scikit_recommender-0.1.1-cp38-cp38-macosx_11_0_x86_64.whl (393.3 kB view hashes)

Uploaded Sep 29, 2023 CPython 3.8 macOS 11.0+ x86-64

scikit_recommender-0.1.1-cp37-cp37m-win_amd64.whl (392.4 kB view hashes)

Uploaded Sep 29, 2023 CPython 3.7m Windows x86-64

scikit_recommender-0.1.1-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (2.4 MB view hashes)

Uploaded Sep 29, 2023 CPython 3.7m manylinux: glibc 2.17+ x86-64

scikit_recommender-0.1.1-cp37-cp37m-macosx_11_0_x86_64.whl (397.8 kB view hashes)

Uploaded Sep 29, 2023 CPython 3.7m macOS 11.0+ x86-64

scikit_recommender-0.1.1-cp36-cp36m-win_amd64.whl (378.1 kB view hashes)

Uploaded Sep 29, 2023 CPython 3.6m Windows x86-64

scikit_recommender-0.1.1-cp36-cp36m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (2.4 MB view hashes)

Uploaded Sep 29, 2023 CPython 3.6m manylinux: glibc 2.17+ x86-64

scikit_recommender-0.1.1-cp36-cp36m-macosx_10_14_x86_64.whl (381.0 kB view hashes)

Uploaded Sep 29, 2023 CPython 3.6m macOS 10.14+ x86-64

Hashes for scikit_recommender-0.1.1-cp311-cp311-win_amd64.whl

Hashes for scikit_recommender-0.1.1-cp311-cp311-win_amd64.whl
Algorithm	Hash digest
SHA256	`ea3cf74edea92829f4a67137efd297c8b684392d687e040107f22d8fff476bdd`
MD5	`85ebee73f05687bfbc8088ec8d67ce6c`
BLAKE2b-256	`003c3142dab70fc8a4ec1c4bb38a26defca70f90f0dcf0364922b0a786748d1d`

Hashes for scikit_recommender-0.1.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl

Hashes for scikit_recommender-0.1.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
Algorithm	Hash digest
SHA256	`c723aebd18e8e7a5482e1018c76344d4efc67e2cdf8350f11b81bc833873dc0c`
MD5	`0357132a830a079c35f65bb76f2edc87`
BLAKE2b-256	`131a0c20c30d7294f9154b32b199567bef9e1271173b7e471dfcf1b9ad8dd8e7`

Hashes for scikit_recommender-0.1.1-cp311-cp311-macosx_10_9_universal2.whl

Hashes for scikit_recommender-0.1.1-cp311-cp311-macosx_10_9_universal2.whl
Algorithm	Hash digest
SHA256	`80c90682954a8c534a87650adca4f1b7cc376117163ca552ab3c6a4839c3fd84`
MD5	`36a39bb7707fcd4252bf95568032b4ed`
BLAKE2b-256	`08c294ed36a7fa176fac1667d7bbef6304fea89934130b6b94a0ed61706e31db`

Hashes for scikit_recommender-0.1.1-cp310-cp310-win_amd64.whl

Hashes for scikit_recommender-0.1.1-cp310-cp310-win_amd64.whl
Algorithm	Hash digest
SHA256	`7affb44a5f6fb48abf2ec64a976da19d0bd48c7e24439a70d800fdf5fe812404`
MD5	`98b7444758d4a2506fb7fd7f8a086c71`
BLAKE2b-256	`e97bd21375862189ef84accd88cdde352c80a0e41373d6cc225baa19224d35e0`

Hashes for scikit_recommender-0.1.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl

Hashes for scikit_recommender-0.1.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
Algorithm	Hash digest
SHA256	`bf745ccf771f5a116f01d1392fb4917d87e4f6de5706c276d342416250fb7fb5`
MD5	`db74ec006b6beccbb1bda37028033ffa`
BLAKE2b-256	`3623117e5e9108a4c5a2c7f0e9570c598836bb29538c4286c55132a216acb45e`

Hashes for scikit_recommender-0.1.1-cp310-cp310-macosx_11_0_x86_64.whl

Hashes for scikit_recommender-0.1.1-cp310-cp310-macosx_11_0_x86_64.whl
Algorithm	Hash digest
SHA256	`b73e83827d308d2091bd1f8f5bb96f78c9dd91a7942727769f036c5d48a7e978`
MD5	`7e2f8f75c89bda5020c8e40e64a12762`
BLAKE2b-256	`b83cab98fa2854df8a2b92351aa8b0d77b8771f1cb720683ce1af7140d969625`

Hashes for scikit_recommender-0.1.1-cp39-cp39-win_amd64.whl

Hashes for scikit_recommender-0.1.1-cp39-cp39-win_amd64.whl
Algorithm	Hash digest
SHA256	`2a31ff71e0742ce77cf1b5f96a36e6da4de9cba1fff0d047bbc4542cb54d170c`
MD5	`dc1c7cbfb6c3448719b06e1a469f46b2`
BLAKE2b-256	`cef0a08cf4401ce8bf32b068b5804e3086126663d2f002e61ba39736f83ee5c9`

Hashes for scikit_recommender-0.1.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl

Hashes for scikit_recommender-0.1.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
Algorithm	Hash digest
SHA256	`66979358abc0b5b865a7e36bbf0bbeee874c919ce7e1168a27d29a2a5288f495`
MD5	`e2b414e6b8bbda8db68e9dfbf26a23dd`
BLAKE2b-256	`f4bee9373a49c8a4a8a124e8a4610c68e4bcfee9c756deee23c2521e882f01cf`

Hashes for scikit_recommender-0.1.1-cp39-cp39-macosx_11_0_x86_64.whl

Hashes for scikit_recommender-0.1.1-cp39-cp39-macosx_11_0_x86_64.whl
Algorithm	Hash digest
SHA256	`21d8bee40590e59bfdbb675ca9c6a6f39fb978f57e278690982098bfbf168736`
MD5	`b99137944df8694c29e59977b282c633`
BLAKE2b-256	`afe4eabcaf9498924256fd271a9ce02784798f8b775fab3763f14487f2d8bac3`

Hashes for scikit_recommender-0.1.1-cp38-cp38-win_amd64.whl

Hashes for scikit_recommender-0.1.1-cp38-cp38-win_amd64.whl
Algorithm	Hash digest
SHA256	`e09f6de77a9e3518cf453f3ada087a407d5eceb765bd0d6bebf28f7c1d2b7b25`
MD5	`7d9948ebb9f61797d08074e1b17cafb6`
BLAKE2b-256	`5ec2f34c50fa985e45cce7a49a4db94135222804ee495acbe416747c41d01cf9`

Hashes for scikit_recommender-0.1.1-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl

Hashes for scikit_recommender-0.1.1-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
Algorithm	Hash digest
SHA256	`538c2034e2f1dafe862d900d3df0501924d72e4507abf8c98d20c77e1e50478f`
MD5	`3001c07f2068a12ef0b8716f19efb0cc`
BLAKE2b-256	`fb1c26dcda953861e4b713c6d2c7d65e2eea8e8c6d5ae1a91cf4476e026aa869`

Hashes for scikit_recommender-0.1.1-cp38-cp38-macosx_11_0_x86_64.whl

Hashes for scikit_recommender-0.1.1-cp38-cp38-macosx_11_0_x86_64.whl
Algorithm	Hash digest
SHA256	`483917cfd2e88ba530f3f12460cb15fc4bb15cc7d3c2f7c6c3559c9ea6b819be`
MD5	`ab358fa70d05c2a327a8c9b06b982172`
BLAKE2b-256	`91052b5e01a5b6b293a9741269f4ccab12f8aae63238cbf180de587eca0eedce`

Hashes for scikit_recommender-0.1.1-cp37-cp37m-win_amd64.whl

Hashes for scikit_recommender-0.1.1-cp37-cp37m-win_amd64.whl
Algorithm	Hash digest
SHA256	`7a11cc6c0108ad2841d40727d65708d15a632ef77518d1b50cbdf2faaa767515`
MD5	`76c3dcfb02a053a7fbede7a086271baf`
BLAKE2b-256	`d4be0e0b86ce9d546dd3178f70ef372c0c8cca5c9f4e8614e89bf2c82243045b`

Hashes for scikit_recommender-0.1.1-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl

Hashes for scikit_recommender-0.1.1-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
Algorithm	Hash digest
SHA256	`989483270af2e50ad35b28756d55923a10cacbf5729ddd5aa5972b123eb786fe`
MD5	`7c3f67092ba04b371146735c2c0970ab`
BLAKE2b-256	`c6212d97e239eea4ae2f08b900b74aeed041c4423ad47d4ea7c9c0a89b59c155`

Hashes for scikit_recommender-0.1.1-cp37-cp37m-macosx_11_0_x86_64.whl

Hashes for scikit_recommender-0.1.1-cp37-cp37m-macosx_11_0_x86_64.whl
Algorithm	Hash digest
SHA256	`4c11840dd82d7c4859eb04d733f642bc48ec1716a055f2f63a08c78f1142b4ed`
MD5	`846eacceba179b62c65e405e8923eeee`
BLAKE2b-256	`fdf3dc74f2f540ffe279e6c1cedd72a471c3993266f16c64003e76134d2f9a98`

Hashes for scikit_recommender-0.1.1-cp36-cp36m-win_amd64.whl

Hashes for scikit_recommender-0.1.1-cp36-cp36m-win_amd64.whl
Algorithm	Hash digest
SHA256	`5feac6c05c6f5911279231c455f8adea713b4c93d0def75c4d0fdeb2a31678c3`
MD5	`a8b42509665fa7429e6a82eff5ab1c31`
BLAKE2b-256	`67655a304ea6581302d69b693bcbdb52d5fd0f385009768c90730ea4a1238055`

Hashes for scikit_recommender-0.1.1-cp36-cp36m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl

Hashes for scikit_recommender-0.1.1-cp36-cp36m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
Algorithm	Hash digest
SHA256	`c9ff9ed99f36c332324fbc56f9d4125d5d3f5a6df1d41513f0336eef5f548d91`
MD5	`67de780bb510f8500925859cfd7b87cc`
BLAKE2b-256	`edf19dc30d30f03d25d331a9da85663228140f41e293b69551d31144d36b56af`

Hashes for scikit_recommender-0.1.1-cp36-cp36m-macosx_10_14_x86_64.whl

Hashes for scikit_recommender-0.1.1-cp36-cp36m-macosx_10_14_x86_64.whl
Algorithm	Hash digest
SHA256	`70a602cf144747e2052ac6e8c38713a025e01f1152caeabebd8be81572e60df7`
MD5	`b52e4f219bd83f677ba6e667a1c76553`
BLAKE2b-256	`bba0f6056dae9db8b1f41cb72d6017e1f82565d597f10108ca47fa74f297d992`