This is a cross-post from the Harry’s Engineering blog. Introduction Here on the Analytics team at Harry’s, we frequently found ourselves training machine learning models on data coming from the entirety of Harry’s existence (4+ years and counting!) in Jupyter notebooks on our laptops. While perhaps this was a feasible workflow in the earlier days…… Continue reading An On-Demand High-Powered Jupyter Notebook Server
For the last week, I’ve been working on deploying Style Transfer Playground, the Flask app that I built that blends the content of one user-uploaded image with the style of another, letting the user see the algorithmically generated hybrid image one iteration at a time in the web browser. The biggest technical challenge when I built…… Continue reading From content streaming to worker queues
I finally did the right thing and added unit testing to my Flask web application. However, I ran into extreme difficulty with testing the authentication feature. I’m using OAuth for authentication. I understand that the appropriate thing to do when testing a feature that makes calls to an external API is to “mock” or “stub out”…… Continue reading Testing OAuth with patch/mock
As I’ve continued building my web app in Flask, there are two error messages that I’ve been encountering repeatedly: RuntimeError: working outside of application context and RuntimeError: working outside of request context I decided to dig a little deeper into what these messages were telling me about Flask’s internals and came across this great Stack Overflow thread (finding…… Continue reading Intro to Flask contexts
I’m working on building a web application with Flask that will algorithmically generate a neural artwork image based on user-uploaded input images and allow the user to see the images generated at intermediate stages of the computation. For the purpose of this post, all you need to know about my project is that I have a large…… Continue reading Streaming algorithmically-generated images with Flask
If you enjoyed my previous post on visualizing Picasso works by style similarity, a generalized version of the code used to generate the embeddings is now available on Github! Embed your own collections of images by style in a two-dimensional scatter plot and color points using custom labels.
In an earlier blog post I explained how I came to discover that “style,” as described in the Gatys, et al style transfer paper, has little correlation with what humans think of when they think of a particular artist’s style, and instead really has more to do with color and texture. In this sense, a particular artist…… Continue reading Picasso “styles” visualized with t-SNE
I recently learned about two different flavors of the Word2Vec model for word embeddings using the original paper and this Tensorflow tutorial. The two architectures are known as Skip-gram and Continuous Bag of Words (CBOW). In both models, the idea is to assign words to vectors in a high dimensional space such that words that are similar in…… Continue reading Word2Vec models explained
This project started with the challenge posed by this Kaggle competition. Essentially, I wanted to determine the likelihood that any two pieces of art were produced by the same person. As a starting point, I looked at the now-famous “style transfer” paper. After reading this paper, I learned that its novelty comes from breaking down a piece…… Continue reading What is artistic style?
Welcome to Andrea’s blog of random thoughts on software, math, machine learning, and more. About me: I’m a recent math PhD turned programmer, currently spending 3 months at the Recurse Center, where I get to spend my days pursuing my new programming interests as they arise. I’ve discovered in my first few weeks that the knowledge…… Continue reading Welcome!