Welcome to GraphLearn’s documentation!

Introduction

• Introduction

GraphLearn-Training

• Introduction
  • Architecture
• Install and Setup
  • Install from wheel.
  • Build from source.
    • Install dependent libraries
    • Build
    • Install
    • (Optional) Install TensorFlow
    • (Optional) Install PyTorch，PyG
  • Docker images
• Quick Start
  • Run through the built-in model
  • GraphLearn-Training Interface Usage
  • Develop a GNN model
    • Data preparation
    • Graph construction
    • Graph sampling
    • Model code
• Graph
  • Data Sources
    • Data Formats
    • Data Source
    • User API
  • Graph object
    • Declaring Graph objects
    • Describing topology
    • Initialization
  • Data objects
    • Dense data objects
    • Sparse data object
  • Graph Operator
  • Graph Operators
    • Graph query
    • Graph traversal
    • Graph Sampling
    • Negative sampling
    • Negative sampling by specified attribute condition
    • KNN
  • Graph Sampling Language (GSL)
    • Introduction
    • GSL syntax
    • SOURCE
    • STEP graph sampling/negative sampling
    • Multiplex downstream
    • Alias
    • SINK
    • GSL execution results
  • Example
• GNN Models
  • Paradigms and Processes
    • Model Paradigm
    • Pipeline
  • Tensorflow
    • Data Layer
    • Feature Processing
    • Model layers
    • Common Losses
    • Global Configuration
    • Examples
    • Kubernets examples
    • Development Guide
    • Other graph learning models
  • PyTorch
    • Development process
• Developer Doc
  • Defining Your Own Operator
    • Programming Interface
    • Distributed Runtime Design
    • Implement a New Operator
    • How to Use an Operator
  • Integrating Other Data Sources
    • File System Interface
    • Register with Scheme
  • System global settings

Dynamic-Graph-Service

• Introduction
  • Architecture
• Service Deployment
  • Prerequisites
  • Deploy kafka Queue Service
  • Installing the Chart
  • Uninstalling the Chart
  • Parameters
• Loading From Your Data Source
  • Rules to Follow
  • Dataloader SDK
  • Data Loading Barrier
• Tutorial
  • 1. Prepare data
  • 2. Train model offline
  • 3. Export TF SavedModel
  • 4. Deploy TF Model
  • 5. Deploy Dynamic Graph Service.
  • 6. Sample and Predict
• JAVA Client apis

Indices and tables

• Index

• Module Index

• Search Page