# `ExBurn.Defn.Compiler`
[🔗](https://github.com/ohhi-vn/ex_burn/blob/main/lib/ex_burn/defn_compiler.ex#L1)

Custom `Nx.Defn.Compiler` that compiles defn expressions
to run on the Burn GPU backend via NIF.

This compiler traverses the `Nx.Defn.Expr` expression tree,
converts it into Burn tensor operations, and executes them
through the Rust NIF layer for GPU acceleration.

## Usage

    Nx.Defn.global_default_options(compiler: ExBurn.Defn.Compiler)

Or per-function:

    defn my_fun(x, y, opts \\ []) do
      Nx.add(x, y)
    end
    compiler: ExBurn.Defn.Compiler

## How it works

1. `Nx.Defn` traces the function body into an expression tree
   of `Nx.Defn.Expr` nodes.
2. The compiler receives the expression tree via `__compile__/4`
   or `__jit__/5`.
3. Each expression node is evaluated by converting parameters
   to Burn tensors, then dispatching the operation to the NIF.
4. The result is converted back to `Nx.Tensor` for the caller.

## Options

  * `:garbage_collect` - when `true`, garbage collects after
    evaluating each node (default: `false`)

---

*Consult [api-reference.md](api-reference.md) for complete listing*