5 ms·

I think that MLops somewhat subdues the modelling and research effort in commercial settings. At least anecdotally I only seem to see the coexistence of ML pipe

I think that MLops somewhat subdues the modelling and research effort in commercial settings. At least anecdotally I only seem to see the coexistence of ML pipelines paired with the “feature engineering” + fit(X,y) approach to data “science”. What I don’t see is first principles research, leading to insight, leading to models to exploit insight which ofttimes requires heavily augmenting how and what data is collected: that part almost never fits neatly anywhere. In its place DS often starts with data then looks for models to fit it. MLops reinforce that pattern, and we end up with the old joke: Man searching for his keys under a street lamp in the dark. Police man asks “Is this where you dropped your keys?”. Man says “No, but this is where the light is”.

Fiahil 4y ago
MLOps is about enabling reproducibility of previous outputs; it has nothing to do with reinforcing any pattern. It's literally "DevOps for Data Science". It's also a sane set of best practices to follow even if you're working on an optimisation case. Maybe you're conflating this with something else ?
- usgroup 4y ago
  > MLOps is about enabling reproducibility of previous outputs; it has nothing to do with reinforcing any pattern. This sentence seems to contradict itself. Otherwise, I'm familiar with MLOps :-) Wikipedia describes it as the intersection of ML, devops and data engineering, which is also how I've experienced it. What data is available (already "plugged in") paints a path of least resistance for a data scientist (DS) who then goes on to look for solutions within that data (see joke above) -- not necessarily, but it tends to go like that. The DS cuts a black box which MLOps tunes, monitors and re-trains: more path of least resistance. This ends up reflecting in hiring decisions, and so on. Again, not necessarily the case, just a dominant pattern I see. I suspect part of the reason is the movement of devs into MLEng and MLOps and analysts/devs into DS because analysts/devs typically do not have research backgrounds. It causes DS -- which initially was almost exclusively PhDs -- to be less about describing phenomena using maths and more about fitting black boxes. Further, "ML" as taught in Google courses (centered on Tensorflow) targets devs and literally paints the process of modelling as features, topologies and parameters. There is a selection pressure in that direction -- presumably due to labour supply, IMO to the detriment of the craft, but hey-ho.

Fiahil 4y ago

MLOps is about enabling reproducibility of previous outputs; it has nothing to do with reinforcing any pattern. It's literally "DevOps for Data Science". It's also a sane set of best practices to follow even if you're working on an optimisation case. Maybe you're conflating this with something else ?

usgroup 4y ago
> MLOps is about enabling reproducibility of previous outputs; it has nothing to do with reinforcing any pattern. This sentence seems to contradict itself. Otherwise, I'm familiar with MLOps :-) Wikipedia describes it as the intersection of ML, devops and data engineering, which is also how I've experienced it. What data is available (already "plugged in") paints a path of least resistance for a data scientist (DS) who then goes on to look for solutions within that data (see joke above) -- not necessarily, but it tends to go like that. The DS cuts a black box which MLOps tunes, monitors and re-trains: more path of least resistance. This ends up reflecting in hiring decisions, and so on. Again, not necessarily the case, just a dominant pattern I see. I suspect part of the reason is the movement of devs into MLEng and MLOps and analysts/devs into DS because analysts/devs typically do not have research backgrounds. It causes DS -- which initially was almost exclusively PhDs -- to be less about describing phenomena using maths and more about fitting black boxes. Further, "ML" as taught in Google courses (centered on Tensorflow) targets devs and literally paints the process of modelling as features, topologies and parameters. There is a selection pressure in that direction -- presumably due to labour supply, IMO to the detriment of the craft, but hey-ho.