Build your machine learning env once for all.

docker setup

On Mac:

1. install docker for mac
2. pull kaggle/python image. http://blog.kaggle.com/2016/02/05/how-to-get-started-with-data-science-in-containers/

pycharm setup

1. add remote env. https://www.jetbrains.com/help/pycharm/configuring-remote-interpreters-via-docker.html
2. API URL: tcp://localhost:3375. (Took me tons of …

When you want to train your neuron nets but don't have Nvidia GPU, you might then need to train it on cloud. FloydHub is a relatively easy and cheap way to do this. Although the document of FloydHub is already intuitive enough, I will go through the setup quickly here …

jupyter notebook

display multiple outputs

from IPython.core.interactiveshell import InteractiveShell
InteractiveShell.ast_node_interactivity = "all"

cell with larger width

from IPython.core.display import display, HTML
display(HTML("<style>.container { width:90% !important; }</style>"))