I want to do data analysis and machine learning for a while, but the environment preparation is a little troublesome, I'm doing hands-on, but I think it's possible that it takes a lot of time just to explain the environment setup.
Jupyter Notebook is a great functional environment, and you can use it with a web browser. It is convenient to put such an environment in Docker, and if you start it with your own Local or start it with a cloud server, anyone with a web browser can use it, so it is easy.
I have a Docker Container that has already been released, but I'm missing some of them, so I made an additional one, so I'll share it.
The Docker file is here. https://github.com/mokemokechicken/jupyter-tensorflow
Contents of https://github.com/jupyter/docker-stacks/tree/master/datascience-notebook
* Jupyter Notebook 4.1.x
* Conda Python 3.x and Python 2.7.x environments
* pandas, matplotlib, scipy, seaborn, scikit-learn, scikit-image, sympy, cython, patsy, statsmodel, cloudpickle, dill, numba, bokeh pre-installed
* Conda R v3.2.x and channel
* plyr, devtools, dplyr, ggplot2, tidyr, shiny, rmarkdown, forecast, stringr, rsqlite, reshape2, nycflights13, caret, rcurl, and randomforest pre-installed
* Julia v0.3.x with Gadfly and RDatasets pre-installed
tensorflow-0.7.1 (python2, python3), skflow
pydot2, pygraphviz
notebooks / under the current directory of the host machineIf you want to move it like this, it will be as follows.
run.sh
docker run -d -p 8888:8888 -v `pwd`/notebooks:/home/jovyan/work -e PASSWORD="$PASSWORD" -e GRANT_SUDO=yes --user root mokemokechicken/jupyter-tensorflow start-notebook.sh
The Option is basically the same as https://github.com/jupyter/docker-stacks/tree/master/datascience-notebook, so please refer to that as well.
I think it would be convenient to deploy one such thing in the company for sharing a little work or studying.