Friday 11 August 2017

How to Build a Data Science Pipeline by Balázs Kégl via @TDataScience

Start with y. Concentrate on formalising the predictive problem, building the workflow, and turning it into production rather than optimising your predictive model. Once the former is done, the latter is easy.

Interesting post which is more useful than the initial description of it suggests.

No comments:

Post a Comment

Note: only a member of this blog may post a comment.