OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Developer Tools & SDKs

Recbole

Recbole is an open-source Python library built on PyTorch for reproducing, developing and evaluating recommendation algorithms in a unified research framework.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Recbole — OpenViST project profile in AI Developer Tools & SDKs.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Recbole for recommendation-system research or development when you want a shared data format, benchmark datasets and evaluation protocols for model experiments.

Who it is for

  • Researchers reproducing or comparing recommendation algorithms.
  • Developers prototyping recommenders with benchmark data and evaluation protocols.

Documented uses

  • Train and evaluate collaborative-filtering or general recommendation models.
  • Experiment with sequential, context-aware or knowledge-based recommendation.
  • Explore click-through-rate prediction, graph neural network and knowledge-graph recommendation approaches.
Before you choose

Requirements & limits

The repository identifies the project as open source and lists an MIT license. The README documents installation through conda, pip or source.

  • The documented environment requires Python 3.7 or later and PyTorch 1.7.0 or later.
  • The quick-start provides a concrete BPR and ml-100k baseline; other models and parameters can be selected through command-line arguments.
  • GPU execution requires CUDA or cudatoolkit 9.2 or later and NVIDIA driver version 396.26 or later on Linux, or 397.44 or later on Windows 10.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Run the documented BPR baseline

Use the quick-start script to train and evaluate BPR on the ml-100k dataset.

Before you start

  • Python 3.7 or later.
  • PyTorch 1.7.0 or later.
  • Linux, Windows 10 or macOS X.

Steps

  1. Install RecBole from the project source using the documented installation instructions.

  2. From the source code, run the command below.

    Documented command
    python run_recbole.py

Expected outcome

The script runs BPR on ml-100k and outputs training progress and validation and test metrics.

Guide limitations

  • The quick-start command uses the default BPR model and ml-100k dataset.
  • GPU use requires CUDA or cudatoolkit 9.2 or later and a sufficiently recent NVIDIA driver.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
15 October 2026
Original website title
RUCAIBox/RecBole · Official homepage title · retained in its source language
How we review sourcesSuggest a correction