back to home

bytedance / monolith

A Lightweight Recommendation System

9,271 stars
714 forks
22 issues
PythonC++Starlark
Chat with CodebaseArchitecture ScanSecurity AuditExplain Codebase

AI Architecture Analysis

This repository is indexed by RepoMind. By analyzing bytedance/monolith in our AI interface, you can instantly generate complete architecture diagrams, visualize control flows, and perform automated security audits across the entire codebase.

Our Agentic Context Augmented Generation (Agentic CAG) engine loads full source files into context, avoiding the fragmentation of traditional RAG systems. Ask questions about the architecture, dependencies, or specific features to see it in action.

Click here to launch the interactive analysis workspace

Embed this Badge

Showcase RepoMind's analysis directly in your repository's README.

[![Analyzed by RepoMind](https://img.shields.io/badge/Analyzed%20by-RepoMind-4F46E5?style=for-the-badge)](https://repomind-ai.vercel.app/repo/bytedance/monolith)
Preview:Analyzed by RepoMind

Repository Summary (README)

Preview

Monolith

What is it?

Monolith is a deep learning framework for large scale recommendation modeling. It introduces two important features which are crucial for advanced recommendation system:

  • collisionless embedding tables guarantees unique represeantion for different id features
  • real time training captures the latest hotspots and help users to discover new intersts rapidly

Monolith is built on the top of TensorFlow and supports batch/real-time training and serving.

Discussion Group

Join us at Discord

https://discord.gg/QYTDeKxGMX

Quick start

Build from source

Currently, we only support compilation on the Linux.

First, download bazel 3.1.0

wget https://github.com/bazelbuild/bazel/releases/download/3.1.0/bazel-3.1.0-installer-linux-x86_64.sh && \
  chmod +x bazel-3.1.0-installer-linux-x86_64.sh && \
  ./bazel-3.1.0-installer-linux-x86_64.sh && \
  rm bazel-3.1.0-installer-linux-x86_64.sh

Then, prepare a python environment

pip install -U --user pip numpy wheel packaging requests opt_einsum
pip install -U --user keras_preprocessing --no-deps

Finally, you can build any target in the monolith. For example,

bazel run //monolith/native_training:demo --output_filter=IGNORE_LOGS

Demo and tutorials

There are a tutorial in markdown/demo on how to run distributed async training, and few guides on how to use the MonolithModel API here.