RNNCell (DiffSharp)

RNNCell Type

Unit cell of a recurrent neural network. Prefer using the RNN class instead, which can combine RNNCells in multiple layers.

Constructor Description

Constructor	Description
`RNNCell(inputSize, hiddenSize, ?nonlinearity, ?bias, ?checkShapes)` Full Usage: `RNNCell(inputSize, hiddenSize, ?nonlinearity, ?bias, ?checkShapes)` Parameters: inputSize : `int` hiddenSize : `int` ?nonlinearity : `Tensor -> Tensor` ?bias : `bool` ?checkShapes : `bool` Returns: `RNNCell`	inputSize : `int` hiddenSize : `int` ?nonlinearity : `Tensor -> Tensor` ?bias : `bool` ?checkShapes : `bool` Returns: `RNNCell`


                
                  
                    
                      RNNCell(inputSize, hiddenSize, ?nonlinearity, ?bias, ?checkShapes)

Full Usage: RNNCell(inputSize, hiddenSize, ?nonlinearity, ?bias, ?checkShapes)

Parameters:

inputSize : int

hiddenSize : int

?nonlinearity : Tensor -> Tensor

?bias : bool

?checkShapes : bool

Returns: RNNCell

Instance member	Description
`this.forwardSequence input` Full Usage: `this.forwardSequence input` Parameters: input : `Tensor` Returns: `Tensor`	input : `Tensor` Returns: `Tensor`
`this.forwardSequenceWithHidden (input, hidden)` Full Usage: `this.forwardSequenceWithHidden (input, hidden)` Parameters: input : `Tensor` hidden : `Tensor` Returns: `Tensor`	input : `Tensor` hidden : `Tensor` Returns: `Tensor`
`this.forwardWithHidden (input, hidden)` Full Usage: `this.forwardWithHidden (input, hidden)` Parameters: input : `Tensor` hidden : `Tensor` Returns: `Tensor`	input : `Tensor` hidden : `Tensor` Returns: `Tensor`
`this.hiddenSize` Full Usage: `this.hiddenSize` Returns: `int`	Returns: `int`
`this.inputSize` Full Usage: `this.inputSize` Returns: `int`	Returns: `int`
`this.newHidden batchSize` Full Usage: `this.newHidden batchSize` Parameters: batchSize : `int` Returns: `Tensor`	batchSize : `int` Returns: `Tensor`