Visualization Resources

Very popular visualzation websites

• https://playground.tensorflow.org/ [written in TypeScript using d3.js.]

Streamlit vs. Shiny vs. Jupyter

• https://www.datarevenue.com/en-blog/data-dashboarding-streamlit-vs-dash-vs-shiny-vs-voila

Streamlit is a dashboard tool based on Python, while Shiny uses R. Both tools focus on turning data analysis scripts into full, interactive web applications.

Streamlit is a full data dashboarding solution, while Jupyter Notebooks are primarily useful to engineers who want to develop software and visualizations. Engineers use Streamlit to build dashboards for non-technical users, and they use Jupyter Notebooks to develop code and share it with other engineers.

R Shiny

https://statistics.calpoly.edu/shiny

This one is very comprehensive.

What's included:

• Regression
• Inference
• Probability and Randomness
• Distribution Theory and Estimation

https://shiny.rit.albany.edu/stat/

~ 25 visualization apps.

https://bcdudek.net/shinyapps.html

~ 20 visualization apps.

https://shiny.rstudio.com/gallery/radiant.html

Radiant is an R package for statistics and machine learning with a Shiny user interface

Misc

• https://rpsychologist.com/cohend/?CER=20&M0=100
• https://r.tquant.eu/
• https://smac-group.github.io/ds/shiny-web-applications.html
• https://plotly-r.com/linking-views-with-shiny.html
• https://datascience-enthusiast.com/R/shiny_ML.html

Shiny for Python

https://r-craft.org/r-news/get-started-with-shiny-for-python/

Streamlit

There seems not to be a Hub for streamlit apps on the topic of statistics & machine learning.

Misc

• https://towardsdatascience.com/interactive-machine-learning-and-data-visualization-with-streamlit-7108c5032144
• https://bgremoval.streamlit.app/
• https://tweets.streamlit.app/
• https://extras.streamlit.app/Toggle%20button
• https://jojo96-asciigan-asciigan-z6pv4q.streamlit.app/
• https://ammaryh92-activation-functions-app-h67vkx.streamlit.app/
• https://christopherdavisuci-streamlit-ed-grad-desc-ehj4d4.streamlit.app/
• https://harmkenn-python-stat-tools-app-8h7fra.streamlit.app/
• https://mesmith027-streamlit-webapps-mc-pistreamlit-app-l4b15e.streamlit.app/

tensorflow.js

Misc

• https://observablehq.com/@nkreeger/visualizing-ml-training-using-tensorflow-js-and-baseball-d
• https://www.youtube.com/watch?v=pQv_lswBeZw
• https://miles-per-gallon-prediction-tensorflow-js.netlify.app/
• https://medium.com/ailab-telu/learn-and-play-with-tensorflow-js-part-1-94ef8a8c60f3
• https://www.oreilly.com/library/view/hands-on-machine-learning/9781838821739/46aecee0-d7c1-4479-9bde-e1f4de5cfd6e.xhtml
• https://www.tensorflow.org/js/demos

Basically, tensorflow.js is rarely used for machine learning visualization purpose, especially for pedagogical goal.

Jupyter Notebook

Misc

• https://github.com/pierpaolo28/Data-Visualization
• https://github.com/stefmolin/python-data-viz-workshop
• https://github.com/DeltaAnalytics/machine_learning_for_good
• https://github.com/LearnDataSci/articles
• https://github.com/shlokashah/Machine-Learning-and-Data-Visualization
• Machine Learning A-Z