ml-switcheroo 🔄🦘

A Deterministic, Specification-Driven Transpiler for Deep Learning Frameworks.

Interactive Demo

Try it right in your browser! The engine runs entirely client-side using WebAssembly.

Live Transpiler Demo

Client-side WebAssembly Engine

Offline

Initialize the engine to translate code between frameworks locally.

⚡ Initialize Engine

PyTorch JAX (Core) TensorFlow NumPy Keras Apple MLX

Flax NNXPaxML / Praxis

⇄

PyTorch JAX (Core) TensorFlow NumPy Keras Apple MLX

Flax NNXPaxML / Praxis

Load Example: -- Select a Pattern --

import torch import torch.nn as nn class Model(nn.Module): def forward(self, x): return torch.abs(x)

Strict Mode Run Translation

Engine Logs

Waiting for engine...

Documentation

• ml-switcheroo 🔄🦘
  • 🚀 Key Features
  • 🏗️ Architecture
  • 📦 Installation
  • 🛠️ CLI Usage
  • ✅ API Support Matrix
  • 🧠 Advanced Capabilities
  • 🔌 Extensibility
  • License
• Architecture
  • 🏗️ The Semantic Pivot Strategy
  • 🧩 1. The Knowledge Base (Hub & Spoke)
  • ⚡ 2. The Transpilation Engine
  • 🔌 3. Framework Adapters (Traits & Hierarchy)
  • 🤖 4. Discovery & Consensus
  • 🧠 5. Plugin System
• Extending
  • 🏗️ Architecture Overview
  • 🔌 2. Adding a Framework Adapter
  • 🧠 3. Plugin System (Custom Code)
• Maintenance
  • 🔄 The Maintenance Lifecycle
  • ⚡ Quick Start: The Bootstrap Script
  • 🛠️ Phase 1: Ingestion (The Hub)
  • 🔗 Phase 2: Mapping (The Spokes)
  • 👻 Phase 3: Ghost Mode Support
  • ✅ Phase 4: Verification (CI Loop)
  • 📚 Documentation & Web Demo
  • 🗃️ Glossary of Artifacts
• Ideas
  • Advanced Features

Reference

• API Reference
  • injector_fw
    • Classes
    • Module Contents
  • ml_switcheroo
    • Usage
    • Submodules
    • Attributes
    • Classes
    • Functions
    • Package Contents
  • injector_spec
    • Classes
    • Module Contents
  • injector_plugin
    • Attributes
    • Classes
    • Module Contents

Indices and tables

• Index

• Module Index

• Search Page