Click or drag to resize

GraphCudnnRNNParamsToCanonical Method

https://www.emgu.com/wiki/index.php/Emgu_TF
CudnnRNNParamsToCanonical

Namespace:  Emgu.TF
Assembly:  Emgu.TF.Netstandard (in Emgu.TF.Netstandard.dll) Version: 2.6.0.1294
Syntax
public Operation CudnnRNNParamsToCanonical(
	Output num_layers,
	Output num_units,
	Output input_size,
	Output parameters,
	long num_params,
	string rnn_mode = null,
	string input_mode = null,
	string direction = null,
	float dropout = 0f,
	long seed = 0,
	long seed2 = 0,
	string opName = "CudnnRNNParamsToCanonical"
)

Parameters

num_layers
Type: Emgu.TFOutput
Input to the operation.
num_units
Type: Emgu.TFOutput
Input to the operation.
input_size
Type: Emgu.TFOutput
Input to the operation.
parameters
Type: Emgu.TFOutput
Input to the operation.
num_params
Type: SystemInt64
num params
rnn_mode (Optional)
Type: SystemString
rnn mode
input_mode (Optional)
Type: SystemString
input mode
direction (Optional)
Type: SystemString
direction
dropout (Optional)
Type: SystemSingle
dropout
seed (Optional)
Type: SystemInt64
seed
seed2 (Optional)
Type: SystemInt64
seed2
opName (Optional)
Type: SystemString
The name of the operation

Return Value

Type: Operation
The operation, where: [0] weights(type: DtInvalid). [1] biases(type: DtInvalid).
See Also