khalido

test -e "${HOME}/.iterm2_shell_integration.bash" && source "${HOME}/.iterm2_shell_integration.bash"

# Tell ls to be colourful
export CLICOLOR=1
export LSCOLORS=Exfxcxdxbxegedabagacad
 
# Tell grep to highlight matches
export GREP_OPTIONS='--color=auto'

khalido

# Configuration file for ipython.
# saved in ~/ipython/profile_default

# enable powerline in ipython
from powerline.bindings.ipython.since_5 import PowerlinePrompts
c = get_config()
c.TerminalInteractiveShell.simple_prompt = False
c.TerminalInteractiveShell.prompts_class = PowerlinePrompts

khalido

exprs__url = "https://github.com/cs109/2014_data/raw/master/exprs_GSE5859.csv"
sampleinfo_url = "https://github.com/cs109/2014_data/raw/master/sampleinfo_GSE5859.csv"

urls = [exprs__url, sampleinfo_url]
filenames = ["exprs_GSE5859.csv", "sampleinfo_GSE5859.csv"]

for url, filename in zip(urls, filenames):
    r = requests.get(url, stream=True)
    with open("data/"+filename, 'wb') as fd:
        for chunk in r.iter_content(chunk_size=256):

khalido

# see https://docs.python.org/3/library/configparser.html for deets

[twitter]

c_key = secret_secret_something
c_secret = secret_secret_something

a_token = secret_secret_something
a_secret = secret_secret_something

khalido

Prologue

• introductory book designed from a computation first, math second point of view.
• the book uses PyMC3, a python package for all things Bayesian
• work through the book by cloning its repo and working through the juypyter notebooks.
• ask questions on stack exchanges stat site Cross Validated.

Chapter 1: Introduction to Bayesian Methods

• we all update our beliefs as more data rolls in - this is the gist of bayesian inference

khalido

# autoreload linked files 
%load_ext autoreload
%autoreload 2

# the viz stuff
%matplotlib inline
import matplotlib.pyplot as plt	# mpl is the base of all python plotting angst
import seaborn as sns			# put some trainer wheels & lipstick on mpl
from IPython import display		# to display pics and other stuff which markdown doesn't

khalido

import boto3	# Amazon's aws library for python 3
import numpy as np
import matplotlib.pyplot as plt

# create a connection to s3
s3 = boto3.resource('s3', 
                    aws_access_key_id=aws_access_key_id, 
                    aws_secret_access_key=aws_secret_access_key)

# you need a bucket, make one and put its name here

khalido

the basics

Open a terminal

Install Command Line Tools for Xcode - needed for homebrew to run

xcode-select --install

Install Homebrew

khalido

# take a single col from your df and drop the nulls 
s = df['some_col'].dropna()

# this should return an expanded df from the col of dicts:
expanded_df = s.apply(lambda row: pandas.Series(row))

# then merge the two dataframes - pandas will put NaNs for the missing values in the expanded df
pd.concat([data1, expanded_df], axis=1)

khalido

Keybase proof

I hereby claim:

• I am khalido on github.
• I am khalido (https://keybase.io/khalido) on keybase.
• I have a public key whose fingerprint is AFB8 329B D65D AD66 8BD7 AD76 D822 F0BE B9BF 0ED2

To claim this, I am signing this object: