Basic Torch Tutorials for Deep Learning
Torch is a scientific computing framework based on Lua[JIT] with strong CPU and CUDA backends.
This repository will handle basic Torch implementations of FFNN(Feed Forward Neural Network), CNN(Convolutional Neural Network), RNN(Recurrent Neural Network), and fine-tuning pre-trained models from scratch.
This tutorial is based on
- One Titan-X(Pascal) local setting for single-GPU handling
- Two Titan-X(Pascal) server setting for multi-GPU handling
See the installation instruction for a step-by-step installation guide. See the server instruction for server setup.
- Install Torch
- Install cuda-8.0
- Install cudnn v5.1
- Linear Regression
- Logistic Regression
- Feedforward Neural Network
- Convolutional Neural Network - mnist
- Convolutional Neural Network - cifar10
- Recurrent Neural Network
- Fine-tuning pre-trained models
- Wide-Residual-Networks, Best CNN results
- Basic GAN
$ git clone https://github.com/meliketoy/Torch-Tutorial.git
$ luarocks install cutorch
$ luarocks install xlua
$ luarocks install optnet
# Install itorch notebook (Ubuntu)
$ sudo apt-get install libzmq3-dev libssl-dev python-zmq
$ git clone https://github.com/facebook/iTorch.git
$ cd iTorch
$ luarocks make