Okay, we're now in day 2. Yesterday I was kind of wrestling with whatever I could find on ABSA, trying to figure out what is the right approach to get some hands on experience. We settled for building an E2E ABSA model using BERT, which we based on a paper …
I'm interviewing for a NLP Data Science position in a couple of days. Figured it would be fun and focusing to journal about it as I go, could be useful to peeps. Technically I already started a few days ago but now it's for real, already got the technical interview …
I've been continuing on my learning physics quests, currently alternating between electromagnetics and waves. I was reading chapter 2 of French's "Vibrations and Waves" and he talked there about Lissajous curves.
These are 2d curves whose x-y coordinates are parametrized by cosines, ie
Today I'm recreating the learner framework from the FastAI course. It's a flexible and quite powerful abstraction around the optimization of the DNN model, which streamlines the user experience. For example, it will be very easy to add different logging capabilities, learning rate finder etc. It is built during the …
Today we'll recreate the fastai notebook on autoencoders, where we train a vanilla autoencoder in FashionMNIST. Even though the autoencoder was actually doing a pretty bad job, it will be good practice for working with HuggingFace databases, CNNs and autoencoders.
Today we will retrace lesson 13-14's notebook that "builds up" pytorch abstractions from scratch. As a first step we'll rederive everything in hardcore numpy (maybe hardcore should be reserved for C). Then we'll start building the abstractions.
Reading the variational autoencoder chapter from the "Understanding Deep Learning" book (which is available for free!). Not trivial, which is why I never got around to learning it, I guess. There are a lot of moving math parts to figure out. One of them is called "the reparametrization trick". So …
I'm trying to debug a weird issue with the Logseq omnivore plugin where it takes forever to sync and it seemingly creates and deletes pages needlessly.
My first step was to properly setting up a dev env (pnpm dev) which didn't work out of the box, instead of just building …
