# ex_burn v0.1.0 - Table of Contents

> Middle layer between Nx and Burn for mobile ML training

## Pages

- [ExBurn](readme.md)

- Guides
  - [Getting Started with ExBurn](01_getting_started.md)
  - [Training Models with ExBurn](02_training.md)
  - [Mobile Deployment with ExBurn](03_mobile_deployment.md)
  - [Architecture Deep-Dive](04_architecture.md)

## Modules

- [ExBurn](ExBurn.md): ExBurn — Elixir bridge to the [Burn](https://burn.dev) deep learning framework.
- [ExBurn.Application](ExBurn.Application.md): Application callback for ExBurn.
- [ExBurn.Backend](ExBurn.Backend.md): Nx backend implementation that delegates tensor operations to Burn via NIF.
- [ExBurn.BurnBridge](ExBurn.BurnBridge.md): High-level bridge for Burn operations.
- [ExBurn.CubeclBridge](ExBurn.CubeclBridge.md): Bridge to ExCubecl for GPU execution via Burn's CubeCL backend.
- [ExBurn.Model](ExBurn.Model.md): Model definition and training orchestration for ExBurn.
- [ExBurn.NifHelper](ExBurn.NifHelper.md): Helper module that wraps NIF calls and returns {:ok, result} tuples.

- [ExBurn.Serving](ExBurn.Serving.md): Nx.Serving integration for ExBurn.
- [ExBurn.Serving.Server](ExBurn.Serving.Server.md): Nx.Serving callback implementation for ExBurn.
- [ExBurn.Tensor](ExBurn.Tensor.md): Tensor conversion utilities between Nx and Burn formats.
- [ExBurn.Training](ExBurn.Training.md): Training loop implementation for ExBurn models.
- [ExBurn.Training.CheckpointCallback](ExBurn.Training.CheckpointCallback.md): Saves model checkpoints at specified intervals.
- [ExBurn.Training.EarlyStoppingCallback](ExBurn.Training.EarlyStoppingCallback.md): Stops training when validation loss stops improving.
- [ExBurn.Training.LoggingCallback](ExBurn.Training.LoggingCallback.md): Logs training metrics after each epoch.

- Exceptions
  - [ExBurn.Error](ExBurn.Error.md): Structured error type for ExBurn operations.