Choosing between AWS, Streamlit Sharing, and Heroku
At a high level, whenever we are trying to deploy our Streamlit application such that users on the internet can see our applications, what we are really doing is renting a computer owned by someone else (such as Amazon) and giving that computer a set of instructions to start up our application. Choosing which platform to use is difficult to know how to do without either having a background in deploying systems or without trying each option out first, but there are a few heuristics that should help you out. The two most important factors for this decision are the flexibility of the system and the time it takes to get up and running. Note that these two factors directly trade off with one another. If you are using Streamlit Sharing, you cannot say "I want this to run on a macOS, and I want to add two GPUs to this app," and so on, but in return, you get a wildly simple process where you can simply point Streamlit Sharing to...