Jeff Hale discdiver

See if you have Visual Studio Code installed by looking in your apps. If you don't, have it installed, install it from here
Configure Visual Studio Code to be your editor for git. From the command line run: git config --global core.editor "code -w"
Mac users: Set Visual Studio Code to open with code from the command line outside VS Code. Instructions here, press command + shift + p
Select Shell Command: Install 'code' command in PATH. Windows users should have the functionality automatically.
Install the Python extension for Visual Studio Code. Click on the Extensions icon in the left side bar and search for Python. Select and install it.
Make black your autoformatter. In the top menu bar go Code -> Preferences -> Settings and type black. Select Under Python > Formating: Provider select black. You may need to pip install black on the comma

Name, Purpose, and GitHub Stars as of 9/28/20

DoWhy - making causal inference easy (Microsoft) - 2,258
CausalML - suite of uplift modeling and causal inference methods using machine learning algorithms based on recent research (Uber) - 1,322
causal impact - implementation of Google's model with all functionalities fully ported and tested - 312
cfrnet - counterfactual regression (doesn't appear to be actively developed) - 135
Causality - tools for causal analysis using observational (rather than experimental) datasets (doesn't appear to be actively developed) - 804
Causal Discovery Toolbox - causal inference in graphs and pairwise settings - 333

Common Pandas Errors

By Jeff Hale

Pandas Version 1.x.x

See the Source file here: https://github.com/discdiver/pandas_errors

Each error is explained, an example is shown, and then the correct code is shown, if applicable.

If you have other common errors you think would be helpful for others, please leave them in the comments and ping me on Twitter @discdiver.

Basic Git Branching Workflow

There are many potential Git workflows. The following is a basic workflow for someone working alone, using different local branches, pushing directly the master remote branch.

To make a new branch locally:

git checkout -b my_branch_name - create and move to a new branch
Do work.
git status - check on things.
git add . - stage changes to be committed.
git commit -m "my_commit-message" - commit changes.

How to fork, clone, and sync from a GitHub repository.

From the GitHub GUI: Fork the repository (click fork).
From your forked repository in the GitHub GUI: Click clone and copy the URL.
From the command line: in the directory you want to put your local copy of the repository, type git status to make sure you are not already in a local git repository folder.
Then type git clone insert_the_url_from_above
cd into the cloned repository.
git remote -v to see the tracking versions. You should see origin, but no upstream.
From the original repository in the GitHub GUI (not yours): Click clone and copy the URL.
git remote add upstream insert_url_of_original_repository_from_the_step_above - the URL to insert will NOT have your username in it.

If you are on a Mac, substitute command for control. Don't type the + (it means press both keys at once).

Shortcuts when in either command mode (outside the cells) or edit mode (inside a cell):

Shift + Enter run selected cell or cells - if no cells below, insert a code cell below
Ctrl + B toggle hide/show left sidebar
Ctrl + S save and checkpoint
Ctrl + Shift + S save as

Set up and use a Python virtual environment

Python virtual environment management can be tricky, but it's essential for isolating dependencies.

You have several options for virtual environment managers. We provide instructions for uv, venv, and conda below.

uv is relatively new, but it's the fastest and becoming very popular. If you don't already have a Python virtual environment manager, we suggest using it.

	import pandas as pd
	import numpy as np
	import plotly.offline as py
	import plotly.graph_objs as go
	import json

	py.init_notebook_mode(connected=False)

	izip = zip

	apiVersion: apps/v1
	kind: Deployment
	metadata:
	name: my-deployment
	spec:
	replicas: 3
	selector:
	matchLabels:
	app: my-app
	template:

	apiVersion: apps/v1 # Deployments get an "apps/v1" - no longer in beta.
	kind: Deployment # kind is for the type of resource.
	metadata: # metadata provides information about the resource.
	name: my-deployment # We name our Deployment my-deployment.
	#This is the value at metadata.name.
	spec:
	replicas: 3 # Specify we want 3 Pods in our ReplicaSet.
	selector:
	matchLabels: # spec.selector.matchLabels tells the Deployment
	app: my-app # to match Pods with a key - value label pair