SIOP 2022 Machine Learning Friday Seminar

Thanks for attending this seminar! This Github contains additional resources for you along with the code from today's session.

Session Recording

https://vimeo.com/707933519/c6651aeb79

Additional Resources

Useful Packages for R

• Tidyverse (Suite of tools for data wrangling and manipulation)
• Tidymodels (Suite of tools for using ML models within the tidyverse)
• Caret (ML models)
• Psych (statistics commonly used in psychology)
• Swirl (to learn R)

Tidymodeling

There are a ton of great resources for learning tidymodeling. I highly suggest you try out the following to get you started.

• https://www.tidymodels.org/
• https://www.tmwr.org/

Useful Packages for Python

• SKLearn (ML models, this package has excellent documentation as well)
• Pandas (used for data manipulation in python)
• numpy (package that contains mathematical functions needed for ML)

Additional Programming Languages and Analysis Tools

During the seminar, we shared coding examples and details on both R and Python. Here are some other programming languages and tools that are sometimes used for machine learning.

• Excel
• AnalyzeIt - Excel package turn Excel into SPSS
• SPSS
• SPSS Modeler
• SAS
• Julia
• Apache

There are also several IDEs (Integrated Development Environments) that can be helpful for making your experience working with R or python smoother. For R, RStudio is the most common IDE. There are multiple options for Python including Pycharm, Spyder, and VScode. Also programs like Sublime and Notepad++ can also display formatted code.

Notebooks, such as Jupyter notebooks or R markdown are also helpful for displaying code and results and for sharing your results with others.

Vendor Services

Another route that organizations sometimes take is using a vendor that offers drag and drop style machine learning services. These services often come with dashboards and common models pre-built and ready for use.

Common vendors include:

• Microsoft Azure
• IBM Watson
• Amazon Web Services (AWS)

Resources for in depth self-guided study

• Online Courses
  • Python for Data Science and Machine Learning Bootcamp
  • Swirl - R package to learn R
  • Codecademy - Introduction to Python
  • Coursera - Andrew Ng
    • AI for Everyone
    • Machine Learning
    • Deep Learning Specialization
• Websites, podcasts, and books
  • Towards Data Science (website)
  • An Introduction to Statistical Learning with Applications in R (book)
  • Artificial Intelligence: A Modern Approach - Peter Norvig (book)
  • Super Data Science (podcast)
  • Practical AI (podcast)
  • 3Blue1Brown (youtube)
  • Two Minute Papers (youtube)