Following my Capstone, I wanted to see if I could host my song lyric generator as a web application. There are tons of tutorials out there; unfortunately, few of them touch on hosting large models (like the LSTM model I developed for my project) and running them on the servers directly supported by sites likeContinue reading “Designing Python Web Applications”
Author Archives: Ange Olson
Capstone Project: Lyric Generation
This May, I concluded my M.S. in Data Science Program and my Capstone project on song lyric generation. While I doubt any of them are going to be hits (most of the generated songs still require a bit of editing, and saying that there’s evidence of a narrative is a stretch), it taught me aContinue reading “Capstone Project: Lyric Generation”
Webscraping Concert Data
As a side project over my winter break from school, I decided to write a script that would take the legwork out of a chore I do every week: figuring out what DC-area concerts should be on my radar. Checking the websites for the different venues is a task perfect for automation given the repetitionContinue reading “Webscraping Concert Data”
Spotify Time Series Analysis Pt. 2
This is a continuation from my previous post, Spotify Time Series Analysis Pt. 1. While not required reading, it might provide some context for what I’m discussing here! A Deep Dive into Genres and Artists How do by artist/genre listening habits change over time? How are different artists and genres related in my listening history?Continue reading “Spotify Time Series Analysis Pt. 2”
Spotify Time Series Analysis Pt. I
Since April of 2016, I’ve made (almost) monthly playlists on Spotify consisting of the songs I enjoyed listening to that month. I created them as the month went on, meaning they’ve got a couple of cool characteristics: They can be used to see which artist/genres I was listening to together at any given time TheyContinue reading “Spotify Time Series Analysis Pt. I”
Spotify Daily Mix Analysis
I recently got back into analyzing my Spotify data and wanted to compare my listening habits to the general population’s. Additionally, while I could probably have a greater diversity in musical taste, I was curious to see how traits of the music I listened to varied across different segments of what I listen to. EveryContinue reading “Spotify Daily Mix Analysis”
Ride Sharing Algorithm Exercise
Awhile back for an old job interview, I was asked to devise a ride sharing algorithm to pool rides. to do this, I was given a large data set (100,000+ data points) and asked to calculate efficiency gains in terms of how many rides had the potential to be pooled and reduce carbon emissions. InContinue reading “Ride Sharing Algorithm Exercise”
Adventures in Spotipy Part II
The other day, I was thinking to myself that alternative music from the 2010s was sadder, maybe even angstier (if that’s a word), than its 2000s counterpart. Not only that, but it felt more mainstream, top-40-friendly as well. I hypothesize that the growing indie scene is a reason for that, but I have no proof.Continue reading “Adventures in Spotipy Part II”
Adventures in Spotipy
Isn’t it a bummer that Spotify’s Wrapped only comes once a year? Well, turns out they offer something that is nearly as good; Spotify Web API Endpoints. You can get basic information like track name, artist name and track popularity, as well as somewhat more obscure information found in “audio analysis.” For my forthcoming analysis,Continue reading “Adventures in Spotipy”
Going Though “Introduction to visualising spatial data in R,” Section II
I’ve recently found some data on AWDT (average weekday daily traffic) for the Seattle area that’s quite detailed. Seattle has a great source of data that you can find via their DOT webpage, including bicycle rack locations and pedestrian and bicycle traffic. This looks like great, easy to work with data for learning GIS inContinue reading “Going Though “Introduction to visualising spatial data in R,” Section II”
Ange Olson 