Fixed-Sample Training

Some runnable examples train repeatedly on one caller-supplied sample:

1. build a model with TorchLean.nn.withModel,
2. wrap it as a ScalarModuleDef (model + supervised loss),
3. load or synthesize one supervised sample (x, y),
4. run steps optimizer updates on that fixed sample, and
5. either print before/after loss or write a TrainLog curve.

This module provides that loop without tying it to a particular model family.

Scope:

• it trains against one fixed sample supplied by the caller;
• it is model-agnostic: callers supply the loss wrapper and optimizer constructor;
• it is backend-agnostic: callers can use it on CPU or CUDA via API.Runtime.Options.

For dataset-backed training, use the TorchLean.Trainer API exported by NN or the shared model-zoo loader helpers.

structure TorchLean.Trainer.FixedSample.LossPair (α : Type) : Type

Before/after scalar losses for a fixed-sample training run.

• beforeLoss : α
• afterLoss : α

@[implicit_reducible]

instance TorchLean.Trainer.FixedSample.instReprLossPair {α✝ : Type} [Repr α✝] : Repr (LossPair α✝)

def TorchLean.Trainer.FixedSample.instReprLossPair.repr {α✝ : Type} [Repr α✝] : LossPair α✝ → ℕ → Std.Format

def TorchLean.Trainer.FixedSample.steps {α : Type} [Context α] [DecidableEq Spec.Shape] [ToString α] [Runtime.FromFloat α] [Runtime.Autograd.Torch.Internal.CudaBridge.TensorConv α] {σ τ : Spec.Shape} (mkModel : nn.M (nn.Sequential σ τ)) (mkModuleDef : (model : nn.Sequential σ τ) → Runtime.Autograd.TorchLean.ScalarModuleDef (Runtime.Autograd.TorchLean.NN.Seq.paramShapes model) [σ, τ]) (mkOptim : (Float → α) → (paramShapes : List Spec.Shape) → Runtime.Autograd.TorchLean.Optimizer α paramShapes) (cast : Float → α) (opts : Runtime.Autograd.Torch.Options) (sample : Sample.Supervised α σ τ) (steps : ℕ) (cudaMemWatch : ℕ := 0) : IO (LossPair α)

One fixed-sample run for an arbitrary scalar backend.

def TorchLean.Trainer.FixedSample.curveFloat {σ τ : Spec.Shape} (mkModel : nn.M (nn.Sequential σ τ)) (mkModuleDef : (model : nn.Sequential σ τ) → Runtime.Autograd.TorchLean.ScalarModuleDef (Runtime.Autograd.TorchLean.NN.Seq.paramShapes model) [σ, τ]) (mkOptim : (paramShapes : List Spec.Shape) → Runtime.Autograd.TorchLean.Optimizer Float paramShapes) (opts : Runtime.Autograd.Torch.Options) (sample : Sample.Supervised Float σ τ) (steps : ℕ) (cudaMemWatch : ℕ := 0) : IO Runtime.Training.Curve

Fixed-sample run specialized to Float, returning a full per-step curve.