Gradient Dissent: Conversations on AI cover image

Shreya Shankar — Operationalizing Machine Learning

Gradient Dissent: Conversations on AI

CHAPTER

ML Tooling - What's Your Takeaway?

The three V's thing made tools or at least the viability of ML tools make a lot more sense like experiment tracking weights and biases is a great example it really 10 X is the velocity experience within experimentation. I think there's an interesting paper gosh I don't remember off the top of my head this is bad I shouldn't know but they pose that hey exact reproducibility is often just not achievable in a lot of ML settings. The friction of integrating an observability or monitoring tool can be pretty high if you get results but people are not getting results. What matters is getting some percentage wise if you're trying to reproduce a model like I want to get the same accuracy or a similar

00:00
Transcript
Play full episode

Remember Everything You Learn from Podcasts

Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
App store bannerPlay store banner