Tutorial in Python targeted at Epidemiologists. Will discuss the basics of analysis in Python 3

Overview

Python-for-Epidemiologists

Join the chat at https://gitter.im/zEpid/community DOI

This repository is an introduction to epidemiology analyses in Python. Additionally, the tutorials for my library zEpid are hosted here. For more information on zEpid, see GitHub or ReadTheDocs.

The directory of this guide is

  1. Python Basics
  2. Basics of pandas (data management library)
  3. Epidemiology analyses in Python
    1. Basics
    2. Missing data
    3. Causal inference
      1. Time-fixed treatments
      2. Time-varying treatments
    4. Sensitivity analyses

Required packages for tutorial

To complete the tutorial, user must have the following packages installed: numpy, pandas, zepid, matplotlib, statsmodels, lifelines, and sklearn

IDE (Integrated Development Environment)

No IDE is required to complete the tutorial. All files are available in ipynb also known as jupyter notebooks. Code can be either downloaded or copied from the notebooks.

Here are some IDEs I have used in the past (and what I believe to be their advantages and disadvantages

Rodeo

This is the IDE I used for a long time. It is set up like RStudio

Advantages:

Basically RStudio but for Python, decent interface, easy to run line-by-line, easy to visualize plots (although it encourage bad habits)

Disadvantages:

Does not have all the features of RStudio (will delete changes if closed without saving), sucks up a lot of memory, sometimes the auto-complete would stop working if I hit more than 300+ lines of code, the environment tab is not great (don't expect it to open anything like RStudio)

Aside: their website has great tutorials how to run some basic stuff in Python if you are new to analysis in Python https://rodeo.yhat.com/

jupyter notebooks

Designed to be like a lab notebook, or like R markdown. Supports a pseudo-line-by-line concept Good for writing, since it allows for MarkDown. While I know a lot of people like jupyter, I only really use it for examples of code, not my personal programming. I never liked how it had to open via a Web Browser. I would rather have it be separate program. However, all guides were made using this IDE

PyCharm

This is the IDE I currently use

Advantages:

Easily set up virtual environments, interacts natively with Git, supports different file formats with plug-ins (e.g. .md), enforces certain coding conventions, better debug code features, organization of files under the project tab are convenient

Disadvantages:

Not great for running line-by-line code (it can do it, just not as elegantly), little more hardcore (I wouldn't really consider it a beginner's IDE. It requires some knowledge of set-up of Python)

IDLE

Ships with the basic Python 3.x installation. It is very basic and does not support line-by-line. Wouldn't recommend unless you are just starting with Python and don't want to commit to an IDE yet

Spyder

Ships with conda. Not bad but I didn't use it that much (I couldn't get the hang of it). Similarly it is an RStudio copy. Can't say too much since I haven't used it extensively

Basic Introduction to Python

If you have never used Python before, I have created some introductory materials to Python and the data management library I use, pandas. These are basic guides, but they also point to other resources. Please READ ALL OF THE BELOW BEFORE PROCEEDING.

Installing Python

To install, Python 3.x, we can download it directly from: https://www.python.org/downloads/

The installer provides an option to add Python3 to PATH, it is highly recommended you do this, since it allows you to avoid having to do it manually

Open Command Prompt / Terminal. When opened, type python and this should open Python in the same window. From here, you can quit by typing 'quit()' or closing the window. If this does NOT work, make sure your environmental variable was created properly

Installing Python Packages

Packages are what stores Python functions that we will use. These packages are contributed by various members of the community (including me)) and there is a wide array. To be able to download packages, we need to make sure we have an environmental variable created for python. We will discuss how to install packages

Python 3.x conveniently comes with a package manager. Basically it stores all the packages and we can use it to download new ones or update already downloaded ones.

To download a new package: Open Command Prompt/Terminal and use the following code (we will be installing pandas)

pip install pandas

To update a Python package, type the following command into Command Prompt. For example, we will update our pandas package

pip install pandas --upgrade

That concludes the basics. Please review parts 1 and 2 of the tutorials next

Comments
  • Cochran-Mantel-Haenszel

    Cochran-Mantel-Haenszel

    Thank you @pzivich for this amazing resource. Having the Hernan/Robbins causal model code in python is super helpful... g-estimation!

    I have a request... do you have a Cochran-Mantel-Haenszel script? If you get the chance, please, it would be useful to us to have in this repo. Thank you in advance!

    opened by opioiddatalab 2
  • Slight changes in Incidence Rate Ratio

    Slight changes in Incidence Rate Ratio

    Incidence Ratio Rate Paragraph

    • Fixed repetition
    • T1 & T0 are defined the same way. I believe that T0 is the person-time contributed by people NOT treated with ART
    opened by jaimiles23 0
  • Updates for v0.8.0

    Updates for v0.8.0

    Checklist for various notebooks to update with v0.8.0 release (hasn't released yet)

    • [x] IPTW update. Lots of major changes, so notebook needs to be completely overhauled

    • [x] Demonstrate new diagnostic functions for IPTW, g-formula, AIPW, TMLE

    • [x] Demonstrate g-bound argument

    • [x] Remove TMLE machine learning custom models. This is being removed in favor of cross-fitting. Can leave how to apply for now, but add the warning and mention will be cut in v0.9.0

    opened by pzivich 0
  • Notebooks not rendering in GitHub

    Notebooks not rendering in GitHub

    Sometimes GitHub has trouble rendering the notebooks. AFAIK the rendering system is behind the scenes at GitHub. Others have this same problem across repos and it sometimes occurs to me as well.

    If the notebook won't render in GitHub, you can copy the URL to the notebook you want to view and use the following site to view the notebook: https://nbviewer.jupyter.org/

    opened by pzivich 0
  • Replicate

    Replicate "Causal Inference"

    Issue to track progress on implementation of Hernan and Robins "Causal Inference" chapters

    • [x] Chapter 12: Inverse probability weights

    • [x] Chapter 13: Parametric g-formula

    • [x] Chapter 14: G-estimation of structural nested models

    • [x] ~Chapter 16: G-estimation for IV analysis~

    • [ ] Chapter 17: Causal survival analysis

    • [ ] Part III: Time-varying treatments

    ~G-estimation is not currently implemented. I will need to implement these before chapter 14 can be done.~

    Currently there are no plans to replicate Chapter 15 (propensity scores and regression) or Chapter 16 (instrumental variables) since the first method does not require zEpid and I am unfamiliar with the second. Maybe instrumental variables will be added in the future?

    For Chapter 16, I am considering demonstrating the usage of g-estimation instead of two-stage least-squares. Specifically, using the same data as done in Chapter 16 but following Technical Point 16.3

    enhancement 
    opened by pzivich 0
  • Tutorials

    Tutorials

    On the website, create quick tutorials demonstrating each of the implemented estimators, descriptions of how they work, and why you might want to use them. Might be more digestible than the current docs (also better justify why to choose one over the other)

    Reference to base on https://lifelines.readthedocs.io/en/latest/jupyter_notebooks/Proportional%20hazard%20assumption.html https://github.com/CamDavidsonPilon/lifelines/blob/master/docs/jupyter_notebooks/Proportional%20hazard%20assumption.ipynb

    TODO

    • [x] Basic measures

    • [x] splines

    • [x] IPTW: time-fixed treatment

    • [ ] IPTW: stochastic treatment

    • [ ] IPTW: time-varying treatment

    • [x] IPCW

    • [x] IPMW: single variable

    • [ ] IPMW: monotone

    • [ ] IPMW: nonmonotone (to add after implemented)

    • [x] G-formula: time-fixed binary treatment, binary outcome

    • [x] G-formula: time-fixed categorical treatment, binary outcome

    • [ ] G-formula: time-fixed continuous treatment, binary outcome (to add after implemented)

    • [x] G-formula: time-fixed binary treatment, continuous outcome

    • [x] G-formula: Monte Carlo

    • [x] G-formula: Iterative Conditional

    • [x] G-estimation of SNM

    • [x] AIPTW

    • [ ] AIPMW

    • [x] TMLE

    • [x] TMLE: stochastic treatment

    • [ ] LTMLE (to add after implemented)

    • [x] Quantitative bias analysis

    • [x] Functional form assessment

    • [x] Generalizability

    • [ ] Transportability (IPSW, g-transport, AIPSW)

    • [x] Monte Carlo g-formula by-hand (helps to explain underlying process)

    opened by pzivich 1
Releases(v0.8.0)
Owner
Paul Zivich
Epidemiology post-doc working in epidemiologic methods and infectious diseases.
Paul Zivich
Proposal, Tracking and Segmentation (PTS): A Cascaded Network for Video Object Segmentation

Proposal, Tracking and Segmentation (PTS): A Cascaded Network for Video Object Segmentation By Qiang Zhou*, Zilong Huang*, Lichao Huang, Han Shen, Yon

Forest 117 Apr 01, 2022
Serving PyTorch 1.0 Models as a Web Server in C++

Serving PyTorch Models in C++ This repository contains various examples to perform inference using PyTorch C++ API. Run git clone https://github.com/W

Onur Kaplan 223 Jan 04, 2023
This is the code used in the paper "Entity Embeddings of Categorical Variables".

This is the code used in the paper "Entity Embeddings of Categorical Variables". If you want to get the original version of the code used for the Kagg

Cheng Guo 845 Nov 29, 2022
Anonymous implementation of KSL

k-Step Latent (KSL) Implementation of k-Step Latent (KSL) in PyTorch. Representation Learning for Data-Efficient Reinforcement Learning [Paper] Code i

1 Nov 10, 2021
Open-source implementation of Google Vizier for hyper parameters tuning

Advisor Introduction Advisor is the hyper parameters tuning system for black box optimization. It is the open-source implementation of Google Vizier w

tobe 1.5k Jan 04, 2023
An evaluation toolkit for voice conversion models.

Voice-conversion-evaluation An evaluation toolkit for voice conversion models. Sample test pair Generate the metadata for evaluating models. The direc

30 Aug 29, 2022
ServiceX Transformer that converts flat ROOT ntuples into columnwise data

ServiceX_Uproot_Transformer ServiceX Transformer that converts flat ROOT ntuples into columnwise data Usage You can invoke the transformer from the co

Vis 0 Jan 20, 2022
Official repo of the paper "Surface Form Competition: Why the Highest Probability Answer Isn't Always Right"

Surface Form Competition This is the official repo of the paper "Surface Form Competition: Why the Highest Probability Answer Isn't Always Right" We p

Peter West 46 Dec 23, 2022
Here we present the implementation in TensorFlow of our work about liver lesion segmentation accepted in the Machine Learning 4 Health Workshop

Detection-aided liver lesion segmentation Here we present the implementation in TensorFlow of our work about liver lesion segmentation accepted in the

Image Processing Group - BarcelonaTECH - UPC 96 Oct 26, 2022
以孤立语假设和宽度优先搜索为基础,构建了一种多通道堆叠注意力Transformer结构的斗地主ai

ddz-ai 介绍 斗地主是一种扑克游戏。游戏最少由3个玩家进行,用一副54张牌(连鬼牌),其中一方为地主,其余两家为另一方,双方对战,先出完牌的一方获胜。 ddz-ai以孤立语假设和宽度优先搜索为基础,构建了一种多通道堆叠注意力Transformer结构的系统,使其经过大量训练后,能在实际游戏中获

freefuiiismyname 88 May 15, 2022
Pytorch implementation of "Training a 85.4% Top-1 Accuracy Vision Transformer with 56M Parameters on ImageNet"

Token Labeling: Training an 85.4% Top-1 Accuracy Vision Transformer with 56M Parameters on ImageNet (arxiv) This is a Pytorch implementation of our te

蒋子航 383 Dec 27, 2022
UAV-Networks-Routing is a Python simulator for experimenting routing algorithms and mac protocols on unmanned aerial vehicle networks.

UAV-Networks Simulator - Autonomous Networking - A.A. 20/21 UAV-Networks-Routing is a Python simulator for experimenting routing algorithms and mac pr

0 Nov 13, 2021
Multi-label classification of retinal disorders

Multi-label classification of retinal disorders This is a deep learning course project. The goal is to develop a solution, using computer vision techn

Sundeep Bhimireddy 1 Jan 29, 2022
Non-Official Pytorch implementation of "Face Identity Disentanglement via Latent Space Mapping" https://arxiv.org/abs/2005.07728 Using StyleGAN2 instead of StyleGAN

Face Identity Disentanglement via Latent Space Mapping - Implement in pytorch with StyleGAN 2 Description Pytorch implementation of the paper Face Ide

Daniel Roich 58 Dec 24, 2022
A library that can print Python objects in human readable format

objprint A library that can print Python objects in human readable format Install pip install objprint Usage op Use op() (or objprint()) to print obj

319 Dec 25, 2022
The official implementation of CircleNet: Anchor-free Detection with Circle Representation, MICCAI 2030

CircleNet: Anchor-free Detection with Circle Representation The official implementation of CircleNet, MICCAI 2020 [PyTorch] [project page] [MICCAI pap

The Biomedical Data Representation and Learning Lab 45 Nov 18, 2022
Attack on Confidence Estimation algorithm from the paper "Disrupting Deep Uncertainty Estimation Without Harming Accuracy"

Attack on Confidence Estimation (ACE) This repository is the official implementation of "Disrupting Deep Uncertainty Estimation Without Harming Accura

3 Mar 30, 2022
Double pendulum simulator using a symplectic Euler's method and Hamiltonian mechanics

Symplectic Double Pendulum Simulator Double pendulum simulator using a symplectic Euler's method. The program calculates the momentum and position of

Scott Marino 1 Jan 12, 2022
Partial implementation of ODE-GAN technique from the paper Training Generative Adversarial Networks by Solving Ordinary Differential Equations

ODE GAN (Prototype) in PyTorch Partial implementation of ODE-GAN technique from the paper Training Generative Adversarial Networks by Solving Ordinary

Somshubra Majumdar 15 Feb 10, 2022
Official codebase for "B-Pref: Benchmarking Preference-BasedReinforcement Learning" contains scripts to reproduce experiments.

B-Pref Official codebase for B-Pref: Benchmarking Preference-BasedReinforcement Learning contains scripts to reproduce experiments. Install conda env

48 Dec 20, 2022