24 #ifndef __ARM_COMPUTE_NEFULLYCONNECTEDLAYER_H__ 25 #define __ARM_COMPUTE_NEFULLYCONNECTEDLAYER_H__ 80 Tensor _interleave4x4_output;
82 Tensor _transpose1xW_output;
84 bool _transpose_weights;
86 bool _batched_fc_layer;
87 bool _accumulate_biases;
Base class for all functions.
Interface for the im2col reshape kernel.
Interface for NEON tensor.
NEFullyConnectedLayer()
Constructor.
NEON kernel to interleave the elements of a matrix.
NEON kernel which transposes the elements of a matrix.
void configure(const ITensor *input, const ITensor *weights, const ITensor *biases, ITensor *output, bool transpose_weights=true)
Set the input and output tensors.
NEON kernel to add a bias to each row of the input tensor.
NEON kernel which transposes the elements of a matrix in chunks of 1x4 if the input data type is F32 ...
Basic implementation of the tensor interface.
Basic function to compute a Fully Connected layer on NEON.
void run() override
Run the kernels contained in the function.
NEON kernel to multiply two input matrices "A" and "B".