Introduction

This is a VirtualBox VM that is automatically generated using Vagrant.

Machine Learning and Deep Learning packages installed are: Scikit-Learn, NLTK, Keras, TensorFlow and PyTorch. A Vagrant file is used to generate this VM based on Ubuntu 16.04 (xenial64).

Getting Started

• This VM should work in Windows, macOS and Linux (tested in Debian-10 Buster)

• VirtualBox (version 6+) is required

• Download and install Vagrant

• Clone the virtual machine specs with:

git clone \
	https://github.com/f0nzie/vagrant-deeplearning-XE64G20U2R5120JPY383AE2007P8888.git

• To generate the virtual machine, change to the folder where this repo has been cloned and type on your local terminal:

  vagrant up
  

• Browse to Jupyter with: http://127.0.0.1:8888/. Try the different notebooks in there.

• To access the virtual machine console or terminal, type:

  vagrant ssh
  

• When finished, power off the virtual machine with:

  vagrant halt
  

What’s Installed

Anaconda3 Python 3.8.3

• Deep Learning
  • Keras 2.4.3
  • TensorFlow 2.2.0
  • PyTorch 1.7.1
• Machine Learning
  • sklearn 0.23.1
  • nltk 3.5
• Numeric and Data Science
  • numpy 1.18.5
  • scipy 1.4.1
  • matplotlib 3.2.2
  • seaborn 0.10.1
  • pandas 1.0.5
  • bokeh 2.1.1
  • h5py 2.10.0
  • xlsxwriter 1.2.9
  • statsmodels 0.11.1
  • regex 2020.6.8

Jupyter notebook server

• Jupyter notebook server is available at the host’s browser at http://localhost:8888.
• Jupyter auto starts at boot by means of systemd jupyter.service
• Service configuration file at /etc/systemd/system/jupyter.service generated by provision-vagrant.sh script.
• Added PATH /home/vagrant/.pyenv/versions/anaconda3-2020.07/bin to Environment to be able to execute conda commands from Jupyter.

Testing

There are several notebooks to test the deep learning packages:

• test_keras.ipynb
• test_tensorflow.ipynb
• theano_test.ipynb
• test_loop_packages.ipynb: check the versions of the main installed packages

Other

• Disk size: 20 GB
• RAM: 5120 GB
• CPUs: 2
• Network: NAT
• USB: off
• Shared folders: 0

Codes machine name

• XE64: Ubuntu Xenial 64-bit

• G20: disk size 20 GB

• U2: 02 CPUs

• R5120: RAM 5120 MB

• JPY383: Jupyter server with Python 3.8.3

• AE2007: Anaconda3 version anaconda3-2020.07

• P8888: Jupyter port 8888

• Host name: XE64G20U2R5120JPY383AE2007P8888

• VM name: vagrant-XE64G20U2R5120JPY383AE2007P8888

• Short name: XE64G20JPY383P8888

Credits

• Thanks to antonpirker for the original Vagrant box.

• Florian Demmer for the Vagrant provisioning.

Notes

• Using Anaconda3-2020.07 installed with pyenv
• Python version 3.8.3