Skip to content

Latest commit

 

History

History
29 lines (23 loc) · 1.72 KB

README.md

File metadata and controls

29 lines (23 loc) · 1.72 KB

Deep Learning tools for the Data Science Virtual Machine

This repo will help you to leverage deep learning applications using the Microsoft's Data Science Virtual Machine (DSVM). It contains two kinds of resources: installation scripts to easily configure the best deep learning libraries in your DSVM and solution packages to let you start with deep learning applications.

Deep Learning Libraries

The following libraries are installed via a simple script:

  • CNTK: Microsoft's native library for deep learning (already preinstalled in the DSVM). Code in C# and C++.
  • mxnet: Distributed library for deep learning. Code in R, python, C++ and Scala.
  • keras: Popular deep learning library. Code in python.
  • caffe: Deep learning library from Berkeley. Code in C++.
  • torch: Mathematical framework for deep learning. Code in Lua and C.
  • chainer: Deep learning library based in caffe. Code in python.
  • theano: Mathematical framework for deep learning. Code in python

To install them all, just log in in your DSVM (in Linux) and write:

sudo chmod +x -R install.sh config
./install.sh

Solutions

Next there are several solutions that will help you learn and understand deep learning.

  • Character Recognition
  • Image classification CIFAR
  • Image classification ImageNet
  • Implementation of artistic style in images
  • Word prediction from characters
  • Sentiment Analisys
  • General machine learning examples