Parallelizing Neural Networks

Development Environment

C++11

CUDA 9.0

Compile and run the train.cu file:

nvcc --std=c++11 train.cu -o train
./train

To execute only on the CPU, change the last argument of the Train() function in train.cu to true.

Name		Name	Last commit message	Last commit date
Latest commit History 6 Commits
layers		layers
.gitignore		.gitignore
README.md		README.md
error_handling.hpp		error_handling.hpp
ffn.cu		ffn.cu
matrix.cu		matrix.cu
presentation.pdf		presentation.pdf
train.cu		train.cu
utils.cu		utils.cu