Skip to content

Latest commit

 

History

History
153 lines (128 loc) · 8.03 KB

Configuration.md

File metadata and controls

153 lines (128 loc) · 8.03 KB
Raw

Configuration: Complete the following one week before class ~1 hr

Contact Instructors and TAs for any obstacles encountered -- do this via course Slack channel access > email

This file contains instructions for how to install terminal (for command line), jupyter notebook (as part of anaconda, which includes python 3), and R (as well as the R kernel for jupyter notebook) on your personal computer. Instructions are provided based on MacOS or Windows 10. If you do not have access to a MacOS or Windows 10 computer, contact the course instructors.

Mac OS *note to scroll to bottom after completing steps 1-5 ['Confirm installations']

1. install python3 and jupyter using anaconda

jupyterPath <- "/miniconda/envs/env_name/bin"
jupyterPath2 <- paste(Sys.getenv("HOME")
"/miniconda/envs/env_name/bin",sep="")

then continue with the rest of the instructions below

2. install R

3. install xcode-select

  • open a terminal and run xcode-select --install

4. configure R kernel

  • launch R
  • execute these commands
jupyterPath <- "/anaconda3/bin"
jupyterPath2 <- paste(Sys.getenv("HOME"),"/anaconda3/bin",sep="")
Sys.setenv(PATH = paste(Sys.getenv("PATH"),jupyterPath,jupyterPath2,sep=":"))
install.packages(c('repr', 'IRdisplay', 'evaluate', 'crayon', 'pbdZMQ', 'devtools', 'uuid', 'digest'))
devtools::install_github('IRkernel/IRkernel')
IRkernel::installspec()

5. start jupyter

  • launch terminal
  • From the bash prompt, execute jupyter notebook. You should see output like this:
$ jupyter notebook
[I 17:13:34.085 NotebookApp] JupyterLab beta preview extension loaded from /anaconda3/lib/python3.6/site-packages/jupyterlab
[I 17:13:34.085 NotebookApp] JupyterLab application directory is /anaconda3/share/jupyter/lab
[I 17:13:34.091 NotebookApp] Serving notebooks from local directory: /Users/sulab
[I 17:13:34.091 NotebookApp] 0 active kernels
[I 17:13:34.091 NotebookApp] The Jupyter Notebook is running at:
[I 17:13:34.091 NotebookApp] http://localhost:8888/?token=8db8050d83f22c9fa642269693264052b92c4863b715937b
[I 17:13:34.091 NotebookApp] Use Control-C to stop this server and shut down all kernels (twice to skip confirmation).
[C 17:13:34.095 NotebookApp] 
    
    Copy/paste this URL into your browser when you connect for the first time,
    to login with a token:
        http://localhost:8888/?token=8db8050d83f22c9fa642269693264052b92c4863b715937b&token=8db8050d83f22c9fa642269693264052b92c4863b715937b
[I 17:13:34.495 NotebookApp] Accepting one-time-token-authenticated connection from ::1
[I 17:13:38.193 NotebookApp] Creating new notebook in 
[I 17:13:38.200 NotebookApp] Writing notebook-signing key to /Users/sulab/Library/Jupyter/notebook_secret
[I 17:13:38.864 NotebookApp] Kernel started: 3769ef70-39e0-4796-9d8d-fab6fb710baf
[I 17:13:39.366 NotebookApp] Adapting to protocol v5.0 for kernel 3769ef70-39e0-4796-9d8d-fab6fb710baf
[I 17:13:48.603 NotebookApp] Starting buffering for 3769ef70-39e0-4796-9d8d-fab6fb710baf:2ae497a1c403450a8418b96d78d1cd05

  • jupyter should have automatically started in a browser. You should see something like this: Jupyter root view

  • confirm the R kernel is successfully installed. If you see the option for R as in the screenshot below, you are done!
    Note If you encounter an error relating to 'installspec()', try the following site

Jupyer R kernel view

Windows Subsystem for Linux

REMINDER: You must have Windows 10 to be able to use the instructions below. If you do not have access to a Windows 10 or MacOS computer, contact the course instructors.

1. install linux

2. install juypter

  • open an ubuntu window from the Windows start menu
  • run the following commands
sudo apt update
sudo apt upgrade
sudo apt install python3 python3-pip ipython3
pip3 install jupyter

3. install R in linux subsystem

  • execute these commands:
sudo add-apt-repository 'deb https://cloud.r-project.org/bin/linux/ubuntu bionic-cran35/'
curl -sL "http://keyserver.ubuntu.com/pks/lookup?op=get&search=0xE298A3A825C0D65DFD57CBB651716619E084DAB9" | sudo apt-key add
sudo apt update
sudo apt install r-base

4. install httr and git2r dependencies

sudo apt-get install libssl-dev libcurl4-gnutls-dev libxml2-dev

5. configure R kernel for jupyter

  • launch R by typing R at the terminal prompt
  • execute these commands:
install.packages(c('repr','IRdisplay', 'evaluate', 'crayon', 'pbdZMQ', 'devtools', 'uuid', 'digest'), INSTALL_opts = '--no-lock')
devtools::install_github('IRkernel/IRkernel')
IRkernel::installspec()

6. start jupyter

  • From a bash prompt, execute jupyter notebook. You should see output like this:
[I 09:26:38.967 NotebookApp] Serving notebooks from local directory: /usr/local/lib/python3.5/dist-packages
[I 09:26:38.967 NotebookApp] 0 active kernels
[I 09:26:38.967 NotebookApp] The Jupyter Notebook is running at:
[I 09:26:38.968 NotebookApp] http://localhost:8888/?token=4f0ec8297e0be4f8789fdcf422196717a3d9221e76e1a933
[I 09:26:38.968 NotebookApp] Use Control-C to stop this server and shut down all kernels (twice to skip confirmation).
[W 09:26:38.972 NotebookApp] No web browser found: could not locate runnable browser.
[C 09:26:38.972 NotebookApp]

    Copy/paste this URL into your browser when you connect for the first time,
    to login with a token:
        http://localhost:8888/?token=xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

  • launch jupyter by copying the URL shown in your terminal window into a browser. You should see something like this: Jupyter root view

  • confirm the R kernel is successfully installed. If you see the option for R as in the screenshot below, you are done!

Jupyer R kernel view

(optional) install cmder

  • download 'mini' version from http://cmder.net/
  • uncompress and copy folder to C:\Program Files\
  • add C:\Program Files\cmder_mini to your Path environment variable (instructions)
    • you may have to log in/out for cmder to be accessible via your start menu, or just click the cmder file in windows explorer
  • to open a bash window
    • within cmder, press control-T. From the "startup command" menu, choose {WSL::bash} or {bash::ubuntu}
    • click the Start button

Confirm installations

  • Are you able to open identify and open terminal on your computer? Y/N
  • Does a browser automatically open to a Jupyter Notebook page showing your files? Y/N
    Type jupyter notebook --version in your terminal to confirm you have v6.0.3 installed
  • In your Jupyter notebook, are you able to initiate a new Python3 script? Y/N
    Type python --version in your terminal to confirm you have v3.8.3 installed
  • In your Jupyter notebook, are you able to initiate a new R script? Y/N
    Type R --version in your terminal to confirm you have v4.0.2 installed
  • Did you make a Github account in the instance you don't already have one? Y/N
    If not, follow the instructions here don't get caught up in the details after, we'll be covering that :)
  • Do you have access to our TSRI #abcb-2020 Slack channel to ask us troubleshooting questions on any of the above? Y/N