Additional Posts in Data & Analytics Consultants
Enthusiast
Are red states now in competition for close-mindedness and ?????
The measure, SB 2777, would allow the state's public school staff to refuse to "use a student's preferred pronoun when referring to the student if the preferred pronoun is not consistent with the student's biological sex" and protect educators from “civil liability and adverse employment action" for doing so”.
Transgender pronoun bill advances in Tennessee's legislature https://www.nbcnews.com/nbc-out/out-politics-and-policy/transgender-pronoun-bill-advances-tennessees-legislature-rcna26045
Chief
Visual Storyteller
Bowl Leader
Any apps to help pacing while running?
Enthusiast
The family everybody knows 😂😂
Additional Posts (overall)
Mentor
Conversation Starter
New to Fishbowl?
Download the Fishbowl app to
unlock all discussions on Fishbowl.
unlock all discussions on Fishbowl.
I have a good memory, so not really.
I do, however, get extremely lost in the business terms. Starting with "data science", "features", "deep" collaborative filtering, "belief networks", etc... Good god, information retrieval is largely referred to as "NLP" for some reason.
I run into issues with junior team members. I talk about the orthogonality of vectors in our reduced hyperplane, and these people look at me all lost. Somehow, they're experts in principal components analysis and commonly lecture the other kiddos on "cosine similarity" but they haven't a single clue about what they're actually talking about.
I have to agree with ZS. A lot of google and stackoverflow warriors mascharade as "data scientists."
For the first time since school. Spoke to another consultant in the firm about their optimisation problem and was also equally lost
Pretty much yeah. When I did time series forecasting after a long time, was screwed; plus you have clients ask for a very specific type of model they saw at a conference somewhere
I get lost when things get super scoped in.
SVM, RF, LInear / logistic regression, etc - fine.
Gradient boosting XLM... that’s when I’m not sure how to apply
What the hell is XLM?
That supposed to be XLR? Heteroscedastic extended logistic regressions to deal with interval scaled, censored outcomes?
I got lost at “xgboost”. Since when the hell does regularizing weak partitions and adding some parallelization constitute new discussion material?
I’m pretty lost as soon as we start getting into legit operations research optimization problems (e.g. anything that won’t solve due to complexity in a fairly simple algebraic formulation).
My brain at this point in time does not think that way. I’d like to improve that, but it’ll be an uphill battle. That said I do love slicing stuff with JuMP. Super cool library.