Caffe2 | fulltopic.github.io

Caffe2

Template and Meta in shared_ptr

Convolution Operation

Recurrent Operation

A Developer’s Guide to Parallel Computing the GPUs

Deeplearning4j

Majhong Supervised Learning

Majhong Tile Classifier

A3C Implementation

DQN Implementation

RNN Implementation

CNN Implementation

Multi-processor Programming

A Solution for No Copy IPC Based on SHM

Art of Multiprocessor Programming Practices Solution

CUDA

Programming Massively Parallel Processors (Solution)

The solutions have not been verified in GPU. Just a draft.

Pytorch

Kernel Registration

Pytorch AWS Installation

Tensor Internal (CPU)

Reinforcement Learning

Sutton. RL:An Introduction

Go