Thread by @DynamicWebPaige

Your company and your projects can get considerable value out of traditional machine learning methods, *without* giant language models or massively-scaled architectures —

just like you can still exercise, or hike, or rock climb, even if you aren't prepping to conquer Everest.

Also: traditional models are generally cheaper to implement; more straightforward to explain and understand; and easier to maintain.

Giant models are exciting, from a research perspective; & can unlock capabilities. But you don't *have* to use them, if a smaller hammer works. 🤷‍♀️

It is shocking to me that there is still such a wide gap between the deep learning community and the traditional data analysis / PyData community. 😞

We are all just nerds trying to use data to understand the world, and build software systems to answer questions/ solve problems.

PS: if you're part of the PyData contingent, and you're on the fence about trying machine learning because you think you might not have the background, or because it seems too challenging:

We all struggle. You are *more than capable* of building interesting and useful projects.

My biggest 💔 is that all the bright, curious humans who *could* be doing machine learning don't feel empowered, or don't have resources.

❤️📚: amazon.com/Hands-Machine-…

scikit-learn 📄: scikit-learn.org/stable/

XGBoost 🌲: xgboost.readthedocs.io/en/latest/

JAX 🧠: jax.readthedocs.io/en/latest/