CNN Training Example

Runnable torchlean cnn example. It trains a small convolutional classifier on a prepared CIFAR-10 minibatch.

The reusable model wiring lives in NN.API.Models.Cnn (nn.models.cnn). This file is the runnable wrapper: command-line parsing, dataset selection, multi-epoch loader training, and TrainLog artifact writing.

python3 scripts/datasets/download_example_data.py --cifar10
lake build -R -K cuda=true && lake exe torchlean cnn --cuda --steps 1

def NN.Examples.Models.Vision.Cnn.exeName : String

def NN.Examples.Models.Vision.Cnn.defaultLogJson : System.FilePath

def NN.Examples.Models.Vision.Cnn.batch : ℕ

def NN.Examples.Models.Vision.Cnn.inC : ℕ

def NN.Examples.Models.Vision.Cnn.inH : ℕ

def NN.Examples.Models.Vision.Cnn.inW : ℕ

def NN.Examples.Models.Vision.Cnn.outDim : ℕ

def NN.Examples.Models.Vision.Cnn.cfg : API.nn.models.CnnConfig

@[reducible, inline]

abbrev NN.Examples.Models.Vision.Cnn.σ : Shape

@[reducible, inline]

abbrev NN.Examples.Models.Vision.Cnn.τ : Shape

def NN.Examples.Models.Vision.Cnn.mkModel : API.nn.M (API.nn.Sequential σ τ)

def NN.Examples.Models.Vision.Cnn.loadCifarLoader {α : Type} [API.Semantics.Scalar α] [API.Runtime.Scalar α] (xPath yPath : System.FilePath) (nRows seed : ℕ) : IO (API.Data.BatchLoader α batch RealData.CifarImage RealData.CifarTarget)

def NN.Examples.Models.Vision.Cnn.main (args : List String) : IO UInt32