This week, #StackOverflowKnows syntactic sugar, overfit or nah, and the definition of norm
We’re in the throes of winter here at Stack Overflow headquarters but the Q&A sites are full of hot questions. This week, we’re wondering if syntactic sugar makes code sweeter, wishing upon a galaxy, and digging into two Scala functions that do basically the same thing but at different speeds.
A star or a galaxy?
When I wish upon a star, does it count extra if it’s a galaxy? Because I’ve been trying find the optimal wish pattern without having to wish for more wishes. Genies are such sticklers for rules.
What would I prefer – an over-fitted model or a less accurate model?
Our question asker is wondering if overfitting a model is worth the 1% extra results. It depends on what kind of data you’re modeling. Speech recognition? That 1% may matter. Everything else? Maybe not.
Why are the fast integer types faster than the other integer types?
Can you speed up operations just by putting the word ‘fast’ in the name of the data type? Because I would like to schedule an appointment with the fast_doctor.
Why is the norm of a complex number in the C++ complex library actually the square of the norm?
Turns out Euclid and C++ don’t agree on what ‘norm’ means. I always thought it was a guy who sat at the bar in Cheers.
Why is zipped faster than zip in Scala?
How can similar Scala functions perform the same task at such different speeds? You’d think that they would be equally zippy.
When to use / not use syntactic sugar
How many syntactic sugars do you take in your code? Or do you like your code like your fancy coffee, verbose and full of multiple elements?
> “Can you speed up operations just by putting the word ‘fast’ in the name of the data type?”
No, no, no. It’s a different type with a different size that has performance improvements for the processor.
I know. It is, how do you say, a joke?
Thanks for letting us know, that helps.
Put the question mark in inverted commas”?” or *”?”*