Minibatch MLP training (batching is implicit)

This next-step file shows the intended "PyTorch-like" minibatch path in TorchLean:

1. Keep the dataset per-sample (x : Vec inDim, y : Vec outDim).
2. Use Data.batchLoader to collate minibatches.
3. Write the model and task over the minibatch shape (batch × Vec inDim).
4. Train one step per minibatch.

Run the loader tutorial instead when possible:

• lake exe torchlean data_csv --epochs 1 --batch 5 --dtype float --backend eager

Build this comparison module directly with:

• lake build NN.Examples.Quickstart.MinibatchMlpTrain

Optional flags (tutorial-specific):

• --data-dir PATH (default: NN/Examples/Data)
• --csv PATH (override the CSV file)
• --seed S
• --batch N
• --epochs E

def NN.Examples.Quickstart.MinibatchMLPTrain.inDim : ℕ

def NN.Examples.Quickstart.MinibatchMLPTrain.hidDim : ℕ

def NN.Examples.Quickstart.MinibatchMLPTrain.outDim : ℕ

def NN.Examples.Quickstart.MinibatchMLPTrain.mkModel {batch : ℕ} : API.nn.M (API.nn.Sequential (Shape.Mat batch inDim) (Shape.Mat batch outDim))

Batched MLP 2 -> 8 -> 1.

def NN.Examples.Quickstart.MinibatchMLPTrain.loadDataset (csvPath : System.FilePath) {α : Type} [API.Semantics.Scalar α] [API.Runtime.Scalar α] : IO (Except String (Runtime.Autograd.Train.Dataset (API.sample.Supervised α (Shape.Vec inDim) (Shape.Vec outDim))))

def NN.Examples.Quickstart.MinibatchMLPTrain.main (args : List String) : IO Unit