Hi! I'm Lj Miranda, and welcome to my blog!
In this blog, I write about my interests in natural language processing, MLOps, and games—so grab a cup of coffee and feel free to look around!
With the world inundated with productivity apps, it feels nice to return to the tried and tested pen-and-paper methods. In this blog post, I'll share my analog "tools for thought" and talk about why I prefer them over digital ones.
Have you ever encountered a problem where your model works well in your test set but doesn't perform well in the wild? It's likely because your test set does not reflect the reality of your domain, overestimating our model's performance. In this blog post, I'll discuss alternative ways to split data and examine their effect on named-entity recognition (NER) benchmarks.
This blog post illustrates how spaCy's registry system works by walking through the process of initializing a spancat pipeline. Read this if you're curious about what happens when you run 'spacy train'!