Data Science from Scratch -- Chapter 9
This is the first section that contained information that’s genuinely new to me. This is probably where my real work begins.
Adventures in Sabbatical Land
This is the first section that contained information that’s genuinely new to me. This is probably where my real work begins.
I took a deeper dive into the documentation for categories, and I discovered something that is making me pause for a bit. Apparently, the purpose of categories was to create some sort of folder structure in the website. I was approaching it more as an “advanced tag” situation, where the overall structure was still flat but the categories would form a virtual nesting organization. For example, all of the Data Science from Scratch posts could be found by looking at the categories and tracing the pattern of learning to books to Data_Science_From_Scratch. But this feels very different from having the categories lead to nested folders.
We’re starting to finally get towards some of the meatier content. This section on gradient descent is a pretty good high level overview. He tries to straddle the line between being mathematically dense (too much for the average person) but while remaining mathematically accurate.
I spent another hour or so wrapping my mind around programming under Jekyll. The good news is that I have a much richer understanding of how it works and the various variables that are floating around the system. The bad news is that I didn’t really get any further in how I wanted the category pages to look.
Hypothesis testing is one of those subjects that I feel like I didn’t understand until the third or fourth time I went through it. Unlike the previous couple chapters, which felt like flyovers, this section actually feels like it goes into a fair amount of depth for an introduction to the topic. Rather than touching on a bunch of different ideas, it leans into two specific examples that drive the entire chapter.
This chapter is a super-brief introduction to probability. It’s a lot like the previous one in that if you don’t know anything going in, I don’t really know how much you’ll know coming out. It just seems way too light on detail.
I spent about an hour trying to work through the details of the tag cloud code to think about how I would like to do that with categories. Tye primary difference with the categories is that I want to be able to have it sort out into sub-categories. For example, if I have the category as learning Data_Science_from_Scratch
, I want the page that shows the posts in the learning
category to have all the Data_Science_from_Scratch
posts grouped together.