TorchLean API

Docs HomeGuideExamplesGraphs

NN.Examples.Advanced.TorchIRPyTorch

TorchLean IR to PyTorch #

Tutorial: TorchLean → IR (NN.IR.Graph) → emitted PyTorch code.

Run: lake exe torchlean torch_ir_pytorch --arch linear > exported_model.py lake exe torchlean torch_ir_pytorch --arch mlp > exported_model.py lake exe torchlean torch_ir_pytorch --arch sum > exported_model.py lake exe torchlean torch_ir_pytorch --arch autoencoder > exported_model.py lake exe torchlean torch_ir_pytorch --arch cnn > exported_model.py lake exe torchlean torch_ir_pytorch --arch conv-mlp > exported_model.py lake exe torchlean torch_ir_pytorch --arch mha > exported_model.py lake exe torchlean torch_ir_pytorch --arch mha-mask > exported_model.py lake exe torchlean torch_ir_pytorch --arch transformer > exported_model.py Then: python3 exported_model.py

Architectures #

source
def NN.Examples.Advanced.TorchIRPyTorch.archLinear :
API.nn.M (API.nn.Sequential (Shape.Vec 2) (Shape.Vec 1))
Instances For
    source
    def NN.Examples.Advanced.TorchIRPyTorch.archMLP :
    API.nn.M (API.nn.Sequential (Shape.Vec 2) (Shape.Vec 1))
    Instances For
      source
      def NN.Examples.Advanced.TorchIRPyTorch.archSumReduce :
      API.nn.M (API.nn.Sequential (Tensor.Shape.Vec 4) Spec.Shape.scalar)
      Instances For
        source
        def NN.Examples.Advanced.TorchIRPyTorch.archAutoencoder :
        API.nn.M (API.nn.Sequential (Shape.Vec 3) (Shape.Vec 3))
        Instances For
          source
          def NN.Examples.Advanced.TorchIRPyTorch.archCNN :
          API.nn.M (API.nn.Sequential (Tensor.Shape.Images 1 1 4 4) (Tensor.shapeOfDims [1, 3]))
          Instances For
            source
            def NN.Examples.Advanced.TorchIRPyTorch.archConvMLP :
            API.nn.M (API.nn.Sequential (Tensor.Shape.Images 1 1 3 3) (Tensor.shapeOfDims [1, 1]))
            Instances For
              source
              def NN.Examples.Advanced.TorchIRPyTorch.archMHA :
              API.nn.M (API.nn.Sequential (Tensor.shapeOfDims [1, 4, 8]) (Tensor.shapeOfDims [1, 4, 8]))
              Instances For
                source
                def NN.Examples.Advanced.TorchIRPyTorch.archMHAMask :
                Spec.Tensor Bool (Tensor.Shape.Mat 4 4)
                Instances For
                  source
                  def NN.Examples.Advanced.TorchIRPyTorch.archMHAMasked :
                  API.nn.M (API.nn.Sequential (Tensor.shapeOfDims [1, 4, 8]) (Tensor.shapeOfDims [1, 4, 8]))
                  Instances For
                    source
                    def NN.Examples.Advanced.TorchIRPyTorch.archTransformer :
                    API.nn.M (API.nn.Sequential (Tensor.shapeOfDims [1, 2, 2]) (Tensor.shapeOfDims [1, 2, 2]))
                    Instances For

                      CLI parsing #

                      source
                      def NN.Examples.Advanced.TorchIRPyTorch.usage :
                      String
                      Instances For

                        Export driver #

                        source
                        def NN.Examples.Advanced.TorchIRPyTorch.emitSeq {σ τ : Spec.Shape} (className : String) (model : API.nn.Sequential σ τ) :
                        IO Unit
                        Instances For
                          source
                          def NN.Examples.Advanced.TorchIRPyTorch.main (args : List String) :
                          IO Unit
                          Instances For