Function arrayfire::convolve2_gradient_nn

pub fn convolve2_gradient_nn<T>(
    incoming_grad: &Array<T>, 
    original_signal: &Array<T>, 
    original_filter: &Array<T>, 
    convolved_output: &Array<T>, 
    strides: Dim4, 
    padding: Dim4, 
    dilation: Dim4, 
    grad_type: ConvGradientType
) -> Array<T> where
    T: HasAfEnum + RealFloating, 

Backward pass gradient of 2D convolution

Parameters

• incoming_gradient gradients to be distributed in backwards pass
• original_signal input signal to forward pass of convolution assumed structure of input is ( d0 x d1 x d2 x N )
• original_filter input filter to forward pass of convolution assumed structure of input is ( d0 x d1 x d2 x N )
• convolved_output output from forward pass of convolution
• strides are distance between consecutive elements along each dimension for original convolution
• padding specifies padding width along each dimension for original convolution
• dilation specifies filter dilation along each dimension for original convolution
• grad_type specifies which gradient to return

Return Values

Gradient Array w.r.t input generated from convolve2_nn