Batch Processing

Layer Input

The forward fully-connected layer accepts the input described below. Pass the Input ID as a parameter to the methods that provide input for your algorithm. For more details, see Algorithms.

Input ID

Input

data

Pointer to the tensor of size n1 x ... x nk x ... x np that stores the input data for the forward fully-connected layer. This input can be an object of any class derived from Tensor.

weights

Pointer to the tensor of size n1 x ... x nk - 1 x m x nk + 1 x ... x np that stores a set of weights. This input can be an object of any class derived from Tensor.

biases

Pointer to the tensor of size m that stores a set of biases. This input can be an object of any class derived from Tensor.

mask

Pointer to the tensor of size n1 x ... x nk - 1 x m x nk + 1 x ... x np that holds 1 for the corresponding weights used in computations and 0 for the weights not used in computations. If no mask is provided, the library uses all the weights.

Layer Parameters

For common parameters of neural network layers, see Common Parameters.

In addition to the common parameters, the forward fully-connected layer has the following parameters:

Parameter

Default Value

Description

algorithmFPType

float

The floating-point type that the algorithm uses for intermediate computations. Can be float or double.

method

defaultDense

Performance-oriented computation method, the only method supported by the layer.

nOutputs

Not applicable

Number of layer outputs m. Required to initialize the algorithm.

Layer Output

The forward fully-connected layer calculates the result described below. Pass the Result ID as a parameter to the methods that access the results of your algorithm. For more details, see Algorithms.

Result ID

Result

value

Pointer to the tensor of size nk x m that stores the result of the forward fully-connected layer. This input can be an object of any class derived from Tensor.

resultForBackward

Collection of data needed for the backward fully-connected layer.

Element ID

Element

auxData

Pointer to the tensor of size n1 x ... x nk x ... x np that stores the input data for the forward fully-connected layer. This input can be an object of any class derived from Tensor.

auxWeights

Pointer to the tensor of size n1 x ... x nk - 1 x m x nk + 1 x ... x np that stores a set of weights. This input can be an object of any class derived from Tensor.

auxBiases

Pointer to the tensor of size m that stores a set of biases. This input can be an object of any class derived from Tensor.

auxMask

Pointer to the tensor of size n1 x ... x nk - 1 x m x nk + 1 x ... x np that holds 1 for the corresponding weights used in computations and 0 for the weights not used in computations. If no mask is provided, the library uses all the weights.

For more complete information about compiler optimizations, see our Optimization Notice.
Select sticky button color: 
Orange (only for download buttons)