YH Michael Wang
Latest
Afterglow Timelapse Studio
Afterglow Timelapse Studio is a timelapse workflow project for assembling image sequences into clean, consistent videos with attention to pacing, light changes, and scene continuity.
Esty Plant Upload Tool
I built the Esty Plant Upload Tool to turn a repetitive Etsy plant-listing workflow into a reviewable pipeline: sort photos, generate listing copy, create a draft, upload images, and assign the shop section.
What I Learned About SVD from XCS224 Homework
A short reflection on what SVD means geometrically, why singular values matter, and how truncated SVD gives useful low-dimensional representations.
What I Learned from Stanford XCS236: Deep Generative Models
My key takeaways from Stanford XCS236, from likelihood-based modeling to diffusion and model tradeoffs in practice.
Accelerated Raytracing Kernel in Python
I did some experimental work in April, where I build bindings for Nvidia's raytracing engine PyOptiX to support writing device kernel in python. Original post: https://developer.nv...
Fast Point in Polygon (PiP) Primitive
Point in polygon test is one of the most fundamental building block in ray tracing. Given a point on a surface and a vertices array representing a polygon, is the point in the poly...
PL:Types intro
This is a series of notes taken from the . Types are essential components of a programming language. From practicality, types separates the definition of different categories of da...
PL:HoF and Closure Idioms
This is a series of notes taken from the . From our of higher order functions, functions can be evaluated, stored and passed as arguments just as other variables. Storing a functio...
PL:Higher-order functions and closures
This is a series of notes taken from the by professor Dan Grossman. In this note and next one, I will discuss higher-order functions and their applications. One key feature that di...
Pointer Generator Network
This article introduces the Pointer Generator Network and Coverage Mechanism. After the seminal paper from Sutzkever in 2014 and Bahdanau in 2015, before Transformer shook the grou...
Fairseq Distributed Training Notes
Fairseq distributed training is largely built on top of the distributed training feature provided by . A couple important notes from their tutorial that will be useful: 1. The exam...
Fairseq Transformer, BART (II)
This is a 2 part tutorial for the Fairseq model BART. In the first part I have walked through the details how a Transformer model is built. Please refer to . A BART class is, in es...