Hey developers this is indeed a cool project . I was wondering where can I get codes for convolution operation, regression model, back -propagation algorithm and all those operations that is required. I have downloaded the github folder and now I am wondering on which file to exactly look at to find the exact codes. Actually I plan to contribute to the project but I think I must go through the already built in codes so that I can produce good quality codes.
It’s all in the folder src/operator and src/operator/tensor.
This tutorial might be helpful to you to get a sense of the organization of the operator files and how to contribute operator implementations.
https://mxnet.incubator.apache.org/how_to/add_op_in_backend.html
1 Like
Where can I get the detail implementation of the namespace mxshadow ?
consider
in_data[reg_enum::kLabel].Size()/in_data[reg_enum::kLabel].shape_[0];
Tensor<xpu, 2> out = out_data[reg_enum::kOut].FlatTo2D<xpu, real_t>(s);
Tensor<xpu, 2> grad = in_grad[reg_enum::kData].FlatTo2D<xpu, real_t>(s);
Tensor<xpu, 2> label = in_data[reg_enum::kLabel]
inside regression_output-inl.h, where are these functions ?