Packages

Production ML model serving on the BEAM. Serve PyTorch models faster than Python with pre-compiled graph execution, AOTI compiled inference, and OTP fault tolerance.

Current section

Files

Jump to
extorch lib extorch application.ex
Raw

lib/extorch/application.ex

defmodule ExTorch.Application do
@moduledoc false
use Application
def start(_type, _args) do
children = [
{Registry, [name: ExTorch.Registry.DType, keys: :duplicate]},
{Registry, [name: ExTorch.Registry.Device, keys: :duplicate]},
{DynamicSupervisor, name: ExTorch.ModelSupervisor, strategy: :one_for_one}
]
Supervisor.start_link(children, strategy: :one_for_one)
end
end