Learn Python by Building Data Science Applications

S3-triggered events

So far, we have written a lambda function that's driven by API requests or is scheduled to be triggered. Now let's complete our application's data cycle by adding an S3-triggered task that will compute the medians for each category every time a new dataset is collected—in other words, once the scheduled task is finished.

In order to do so, we'll add one more operation to the same data collection application that we've been working on:

1. First, let's redefine the median computation. The original code also used pandas, so we have two options: use NumPy as we just did for the ML part, or use vanilla Python. As our raw data collection doesn't need NumPy, let's stick with the second option. Here is the no-dependency code for ...