TorchLean API

Docs HomeGuideExamplesGraphs

NN.API.Models.PPO

PPO Model Helpers (API) #

Reusable actor/critic MLP constructors for PPO examples.

These helpers intentionally cover only the neural-network shape. Environment collection, trust boundary checks, advantage computation, and optimizer loops stay in the examples/runtime modules.

source
structure NN.API.nn.models.PPOActorCriticConfig :
Type

Configuration for a simple PPO actor/critic pair over vector observations.

Instances For
    source
    @[implicit_reducible]
    instance NN.API.nn.models.instReprPPOActorCriticConfig :
    Repr PPOActorCriticConfig
    source
    def NN.API.nn.models.instReprPPOActorCriticConfig.repr :
    PPOActorCriticConfig → ℕ → Std.Format
    Instances For
      source
      @[reducible, inline]
      abbrev NN.API.nn.models.ppoActorInShape (cfg : PPOActorCriticConfig) (pfx : Shape) :
      Shape
      Instances For
        source
        @[reducible, inline]
        abbrev NN.API.nn.models.ppoActorOutShape (cfg : PPOActorCriticConfig) (pfx : Shape) :
        Shape
        Instances For
          source
          @[reducible, inline]
          abbrev NN.API.nn.models.ppoCriticOutShape (_cfg : PPOActorCriticConfig) (pfx : Shape) :
          Shape
          Instances For
            source
            def NN.API.nn.models.ppoActor (cfg : PPOActorCriticConfig) (pfx : Shape) :
            M (Sequential (ppoActorInShape cfg pfx) (ppoActorOutShape cfg pfx))

            Actor MLP mapping observations to action logits.

            Instances For
              source
              def NN.API.nn.models.ppoCritic (cfg : PPOActorCriticConfig) (pfx : Shape) :
              M (Sequential (ppoActorInShape cfg pfx) (ppoCriticOutShape cfg pfx))

              Critic MLP mapping observations to a scalar value estimate.

              Instances For