Public | Automated Build

Last pushed: a year ago
Short Description
Basic Data Science Notebook Image. Julia, R and Python
Full Description

Data Science Image

Lightweight image with Julia, R and Python and notebook kernels.

Docker Hub Repo
Github Source

Based off of:

  • Ubuntu 16.04 LTS
  • Tensorflow 0.11 (CPU only)
  • Julia 0.5.0
  • R 3.3.2
  • Python 3.5.2
    • keras
    • numpy
    • pandas
    • scikit-learn
    • xgboost

To start a notebook:

$ docker run -d -p 8888:8888 -v `pwd`:/home/work burrito/notebook sh -c "jupyter notebook --no-browser --port 8888 --ip=0.0.0.0"
$ docker run -d -p 8888:8888 burrito/notebook sh -c "jupyter notebook --no-browser --port 8888 --ip=0.0.0.0"

$ docker run -i -t -p 8888:8888 burrito/notebook
$ docker exec `container ID` jupyter notebook --no-browser --port 8888 --ip=0.0.0.0

docker run -d -p 8888:8888 -v pwd:/home/work burrito/notebook jupyter notebook --no-browser --port 8888 --ip=0.0.0.0
docker run -d -p 8888:8888 burrito/notebook jupyter notebook --no-browser --port 8888 --ip=0.0.0.0
These should work, but https://github.com/ipython/ipython/issues/7062

Find IP address with docker-machine ip default on Windows
Visit http://127.0.0.1:8888/ (with the noted IP) for Jupyter Notebook
Or http://localhost:8888/ on Mac/Linux

To start container interactively:

$ docker run -it burrito/notebook
$ R

To run a Python 3 script:

$ docker run -it -v `pwd`:/home/work burrito/notebook python3 script.py

Copy files in or out:

$ docker cp `local directory` `container ID`:/home/work
$ docker cp `container ID`:/home/work/data.csv `local directory`

Run Tensorflow from R:

library(tensorflow)
sess = tf$Session()
hello <- tf$constant('Hello, TensorFlow!')
sess$run(hello)

Docker Pull Command
Owner
burrito
Source Repository