3 ms·
Why do we have to “ML” everything? A simple database query should provide good evidence. I’m confident that a significant portion of problems that ML is applie
by iraklism 8y ago
Why do we have to “ML” everything? A simple database query should provide good evidence.
I’m confident that a significant portion of problems that ML is applied to today can be solved by some Excel sheets.
- mynameisvlad 8y agoBecause nobody wants to spend time crunching an Excel sheet when you can train a computer to do it for you, automatically, all the time.
- steve19 8y agoMachine learn is also called statistical learning. It does not mean deep learning which is a subset of ML. Creating a model in python or R is going to be much easier than in excel.
- alexgmcm 8y ago>Creating a model in python or R is going to be much easier than in excel. True, but my boss will ask for it in Excel anyway.
- kps 8y ago“Half of the time when companies say they need ‘AI’ what they really need is a SELECT clause with GROUP BY.” — https://twitter.com/matvelloso/status/1001899539484192768 https://twitter.com/matvelloso/status/1001899539484192768
- orasis 8y agoIf you have a strong ML work flow it’s going to be easier to scale and automate an ML solution than an excel solution.