💻 convenience scripts for easily setting up package libraries
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
.gitignore
LICENSE
Makefile
README.md
atompkgs.sh
juliapkgs.jl
jupyter.md
pypkgs-pip.py
rpkgs.R

README.md

myPkgLib MIT license

Package libraries and setup scripts for R, Python, Julia, and Atom (editor)


Directions:

To set up package libraries for R, Python, Julia, and the Atom editor, the following order of steps is recommended:

On Ubuntu (Linux) run as root by adding prefix sudo

  1. R: Rscript -e rpkgs.R (or in an R session)
  2. Python v3.4+: python3 pypkgs.py (on Ubuntu 14.04, use python3.5)
  3. Python v2.7+: python2 pypkgs.py (on Ubuntu 14.04, use python2.7)
  4. Julia: julia juliapkgs.jl
  5. Atom: sh atompkgs.sh

N.B., when install R packages, permissions issues can most easily be avoided by running the install script in an interactive R session started from the terminal (in particular, use of R CMD BATCH is not recommended). Rscript seems to work well (as of 15 June 2016), but reliability has not yet been completely verified.


Caveat Emptor

The following information is current as of 27 August 2016 and should be taken into account when following the recommended install procedure above.

  • On macOS, the "system Python" is Python 2.6.6 and is symlinked as python. After installation of newer versions of Python (e.g., via Homebrew), Python 2.7.X is symlinked as python2, and Python 3.5.X is symlinked as python3.

  • On Ubuntu Trusty Tahr (14.04), the "system Python" is Python 3.4.3 and is symlinked as python3. After installing newer Python versions (via Felix Krull's "deadsnakes" PPA), Python 3.5.X is symlinked as python3.5, while Python 2.7.X is symlinked as Python2.7.


Supplementary Notes

  1. Set up Jupyter kernels for R, Julia, Python, Torch, etc., manually using the instructions in the file jupyter.md. Note that Jupyter kernels for R, Python3, and Julia are set up in the individual language-specific install scripts. The file jupyter.md is updated semi-regularly to facilitate installation of other kernels (e.g., Python2, Torch).

  2. Download and install the pythonic Xonsh shell by following these instructions for Linux or for macOS. Note that xonsh and gitsome are automatically installed for Python v.3.4+ in the pypkgs.py script.


License

© 2016-2018 Nima Hejazi

The contents of this repository are distributed under the MIT license. See file LICENSE for details.