Welcome to Diode’s documentation!

diode is a framework that makes it easy to develop machine learning models that plug into the external interfaces of torch and torch.compile. It allows users to both gather data from torch and train Machine Learning models on the gathered data.

Contents:

• Accelerating torch.compile with Diode
  • Overview
  • Quick Start
  • Complete Example
  • Benefits
  • Advanced Configuration
  • Integration with Existing Workflows
  • Supported Operations
• Getting Started with Training Models with Diode
  • Overview
  • Prerequisites
  • Step 1: Data Collection
  • Step 2: Model Training
  • Step 3: Creating a Validation Dataset
  • Step 4: Model Validation
  • Complete Workflow Script
  • Advanced Configuration
  • Tips
  • Next Steps
• API Documentation
  • diode package

Articles:

• Project Overview
• torch-diode
  • Target Audience:
  • Features:
  • Featured Models
  • Option 1: Installation with Pre-Trained Models
  • Option 2: Installation without Pre-Trained Models
  • Model Organization
  • Get Started
• Contributing Guide
• Contributing to Diode
  • Table of Contents
  • Getting Started
  • Development Setup
  • Project Structure
  • Contributing Guidelines
  • Code Style
  • Testing
  • Continuous Integration (CI)
• Testing and Code Coverage
  • Prerequisites
  • Running Tests
  • Code Coverage
  • Test Markers
  • CI/CD Integration
  • Submitting Changes
  • Community
• Packaging Guide
• Diode Packaging Setup
  • Package Configurations
  • File Inclusion
  • Building the Packages
  • Usage Differences
  • Configuration Files
  • Key Differences
  • Deployment
  • Dependencies
  • File Structure
  • Choosing the Right Package
• Workflows Guide
• Matrix Multiplication Toolkit
  • Overview
  • Usage
  • Key Features
  • Command-Line Options
  • Common Options
  • Data Collection Options
  • Model Training Options
  • Advanced Usage
  • Internal Structure
• Systems Architecture
• Torch-Diode Systems Architecture
  • Key Architecture Components

Reference:

• Index