Nueral Network Workshop – Lab 1 Windows

Install Anaconda, Graphviz, and Python Packages

Click here for macOS Instructions

Install GraphViz

  1. Install https://graphviz.gitlab.io/_pages/Download/Download_windows.html
  2. Add the install folder of graphviz to your system path, by default this is C:\Program Files (x86)\Graphviz2.38\bin (Video Tutorial

[collapse]
Install Anaconda and Python

  1. Install Windows 64 bit Python 3.x https://www.anaconda.com/download/#windows

[collapse]
Install Packages

  1. Open Anaconda Prompt
  1. conda install theano
  2. conda install mingw libpython
  3. conda install pydot
  4. pip install tensorflow
  5. pip install keras
  6. conda update –all
  7. Close Anaconda Prompt

[collapse]
Test Install

Test Install

  1. Open Spyder
  2. Create a New File
  3. Add the following code and run it:

A successful run will use theano and determine if you are using GPU or CPU to do the work (You will be using CPU)

Notice the content of the last three lines in the screenshot!

[collapse]

 

Lab Complete!

 

 

Extra Credit – Configure Nvidia GPU (Cuda) Support

Install Cuda and cuDnn

  1. Install https://developer.nvidia.com/cuda-90-download-archive
  2. Add folder to path
  3. Add bin folder to path (C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v9.0\bin)
  4. Install https://developer.nvidia.com/rdp/cudnn-download
  5. Open Anaconda Prompt
  6. pip install tensorflow-gpu
  7. conda install pygpu
  8. Close Anaconda Prompt
  9. Repeat the install Test above, verify your GPU is now being used

 

[collapse]