Computer Architecture final project

Install:

$ git clone https://github.com/Shin-Yan/NYCU_computer_architecture_final.git

Compile:

// In developing mode, use make
$ make

// In Virtual Machine to test the code, use make vm
$ make vm

Execution: Note that gpu_version2 is just for debugging on your computer, it can't actually access the CUDA device.

// In developing mode
$ ./gpu_version1
$ ./gpu_version2

// In Virtual Machine
$ ./gpu_vm1
$ ./gpu_vm2

topic: Forward propagation matrix multiplication with cuda

Environment

File discription

Algorithm

Code explaination

Experiment result