Safe Haskell	None
Language	GHC2024

MnistRnnRanked2

Description

Ranked tensor-based implementation of Recurrent Neural Network for classification of MNIST digits. Sports 2 hidden layers.

Synopsis

type ADRnnMnistParametersShaped (target :: Target) (width :: Nat) r = (LayerWeigthsRNNShaped target SizeMnistHeight width r, LayerWeigthsRNNShaped target width width r, (target (TKS '[SizeMnistLabel, width] r), target (TKS '[SizeMnistLabel] r)))
type LayerWeigthsRNNShaped (target :: Target) (in_width :: Nat) (out_width :: Nat) r = (target (TKS '[out_width, in_width] r), target (TKS '[out_width, out_width] r), target (TKS '[out_width] r))
type ADRnnMnistParameters (target :: Target) r = (LayerWeigthsRNN target r, LayerWeigthsRNN target r, (target (TKR 2 r), target (TKR 1 r)))
type LayerWeigthsRNN (target :: Target) r = (target (TKR 2 r), target (TKR 2 r), target (TKR 1 r))
zeroStateR :: forall target r (n :: Nat) a. (BaseTensor target, GoodScalar r) => IShR n -> (target (TKR n r) -> a) -> a
unrollLastR :: forall target state c w r (n :: Nat). (BaseTensor target, GoodScalar r, KnownNat n) => (state -> target (TKR n r) -> w -> (c, state)) -> state -> target (TKR (1 + n) r) -> w -> (c, state)
rnnMnistLayerR :: (ADReady target, GoodScalar r, Differentiable r) => target (TKR 2 r) -> target (TKR 2 r) -> LayerWeigthsRNN target r -> target (TKR 2 r)
rnnMnistTwoR :: (ADReady target, GoodScalar r, Differentiable r) => target (TKR 2 r) -> PrimalOf target (TKR 2 r) -> (LayerWeigthsRNN target r, LayerWeigthsRNN target r) -> (target (TKR 2 r), target (TKR 2 r))
rnnMnistZeroR :: (ADReady target, GoodScalar r, Differentiable r) => Int -> PrimalOf target (TKR 3 r) -> ADRnnMnistParameters target r -> target (TKR 2 r)
rnnMnistLossFusedR :: (ADReady target, ADReady (PrimalOf target), GoodScalar r, Differentiable r) => Int -> (PrimalOf target (TKR 3 r), PrimalOf target (TKR 2 r)) -> ADRnnMnistParameters target r -> target ('TKScalar r)
rnnMnistTestR :: (target ~ Concrete, GoodScalar r, Differentiable r) => Int -> MnistDataBatchR r -> ADRnnMnistParameters target r -> r

Documentation

type ADRnnMnistParametersShaped (target :: Target) (width :: Nat) r = (LayerWeigthsRNNShaped target SizeMnistHeight width r, LayerWeigthsRNNShaped target width width r, (target (TKS '[SizeMnistLabel, width] r), target (TKS '[SizeMnistLabel] r))) Source #

The differentiable type of all trainable parameters of this nn. Shaped version, statically checking all dimension widths.

type LayerWeigthsRNNShaped (target :: Target) (in_width :: Nat) (out_width :: Nat) r = (target (TKS '[out_width, in_width] r), target (TKS '[out_width, out_width] r), target (TKS '[out_width] r)) Source #

type ADRnnMnistParameters (target :: Target) r = (LayerWeigthsRNN target r, LayerWeigthsRNN target r, (target (TKR 2 r), target (TKR 1 r))) Source #

The differentiable type of all trainable parameters of this nn.

type LayerWeigthsRNN (target :: Target) r = (target (TKR 2 r), target (TKR 2 r), target (TKR 1 r)) Source #

zeroStateR :: forall target r (n :: Nat) a. (BaseTensor target, GoodScalar r) => IShR n -> (target (TKR n r) -> a) -> a Source #

unrollLastR :: forall target state c w r (n :: Nat). (BaseTensor target, GoodScalar r, KnownNat n) => (state -> target (TKR n r) -> w -> (c, state)) -> state -> target (TKR (1 + n) r) -> w -> (c, state) Source #

rnnMnistLayerR Source #

Arguments

:: (ADReady target, GoodScalar r, Differentiable r)
=> target (TKR 2 r)	in state, `[out_width, batch_size]`
-> target (TKR 2 r)	input, `[in_width, batch_size]`
-> LayerWeigthsRNN target r	parameters
-> target (TKR 2 r)	output state, `[out_width, batch_size]`

A single recurrent layer with tanh activation function.

rnnMnistTwoR :: (ADReady target, GoodScalar r, Differentiable r) => target (TKR 2 r) -> PrimalOf target (TKR 2 r) -> (LayerWeigthsRNN target r, LayerWeigthsRNN target r) -> (target (TKR 2 r), target (TKR 2 r)) Source #

Composition of two recurrent layers.

rnnMnistZeroR Source #

Arguments

:: (ADReady target, GoodScalar r, Differentiable r)
=> Int	batch_size
-> PrimalOf target (TKR 3 r)	input data `[sizeMnistWidth, sizeMnistHeight, batch_size]`
-> ADRnnMnistParameters target r	parameters
-> target (TKR 2 r)	output classification `[SizeMnistLabel, batch_size]`

The two-layer recurrent nn with its state initialized to zero and the result composed with a fully connected layer.

rnnMnistLossFusedR :: (ADReady target, ADReady (PrimalOf target), GoodScalar r, Differentiable r) => Int -> (PrimalOf target (TKR 3 r), PrimalOf target (TKR 2 r)) -> ADRnnMnistParameters target r -> target ('TKScalar r) Source #

The neural network composed with the SoftMax-CrossEntropy loss function.

rnnMnistTestR :: (target ~ Concrete, GoodScalar r, Differentiable r) => Int -> MnistDataBatchR r -> ADRnnMnistParameters target r -> r Source #

A function testing the neural network given testing set of inputs and the trained parameters.