This repository is an extension of Caffe for the paper "Statistical Loss and Analysis for Deep Learning in Hyperspectral Image Classification" (TNNLS). [paper]
-
Install prerequisites for
Caffe
(see: Caffe installation instructions) -
Add the header and source file in the Caffe project.
-
Modify
#include <complex.h>
#define lapack_complex_float float _Complex
as
#include<complex>
#define lapack_complex_float std::complex<float>
in the dependency package NugetPackages/OpenBLAS.0.2.14.1/lib/native/include/lapacke.h
- Compile Caffe framework.
If you find this work helpful for your research, please consider citing:
@article{gong2020,
Author = {Zhiqiang Gong and Ping Zhong and Weidong Hu},
Title = {Statistical Loss and Analysis for Deep Learning in Hyperspectral Image Classification},
Booktitle = {IEEE Transactions on Neural Networks and Learning Systems},
Year = {2020}
}