ReidWilliams

# rename a dataframe column
dataframe = dataframe.rename(index=str, columns={'old_col_name': 'new_col_name'})

# get all possible values for a string / categorical column
dataframe[column].value_counts(dropna=False)

# group data and count number of rows in a group
grouped = dataframe.groupby(color)
nreds = len(grouped.get_group('red'))

ReidWilliams

$ pip freeze > requirements.txt
$ pip install -r requirements.txt

ReidWilliams

# create environment with jupyter
conda create -n ENVNAME python=3 jupyter

ReidWilliams

# Easy way to remotely access a Jupyter notebook, by forwarding a port
# Forward local port 8888 to the remote's localhost port 8888
ssh -L 8888:localhost:8888 user@10.2.2.99

ReidWilliams

while true; do rsync -e "ssh -i /PATH/TO/PRIVATE/KEY.pem" -azvh --delete ./LOCAL-REPO/ ec2-user@IP:/REMOTE/PATH; sleep 2; done

ReidWilliams

kill -9 $(ps -ef | grep '[n]ode --debug-brk=5858' | awk '{print $2}')

ReidWilliams

// Uses underscore.js to wrap a function and upon calling
// prints the function source and the argument.
// Useful for debugging chains of promises.
// USAGE
// var myfunction = function(argument) {...};
// myfunction = inspect(myfunction); // myfunction is now instrumented

var _ = require('underscore');
var util = require('util');
var inspect = function(fnToInspect) {