MCPcopy Create free account
hub / github.com/davisking/dlib / batch_normalize_conv_gradient

Function batch_normalize_conv_gradient

dlib/cuda/tensor_tools.cpp:790–810  ·  view source on GitHub ↗

Source from the content-addressed store, hash-verified

788 }
789
790 void batch_normalize_conv_gradient (
791 const double eps,
792 const tensor& gradient_input,
793 const tensor& means,
794 const tensor& invstds,
795 const tensor& src,
796 const tensor& gamma,
797 tensor& src_grad,
798 tensor& gamma_grad,
799 tensor& beta_grad
800 )
801 {
802
803 IF_DLIB_USE_CUDA(
804 cuda::batch_normalize_conv_gradient(eps,gradient_input, means, invstds, src, gamma, src_grad, gamma_grad, beta_grad);
805 )
806
807 IF_DLIB_NOT_USE_CUDA(
808 cpu::batch_normalize_conv_gradient(eps,gradient_input, means, invstds, src, gamma, src_grad, gamma_grad, beta_grad);
809 )
810 }
811
812// ----------------------------------------------------------------------------------------
813

Callers

nothing calls this directly

Calls

no outgoing calls

Tested by

no test coverage detected