It also shows not-so-nice alternatives
The only implementations for
(==) :: a -> a -> Bool ought to beeq1 :: a -> a -> Bool
eq1 _ _ = False
eq2 :: a -> a -> Bool
eq2 _ _ = True
apart from silly ones like
eq1 = undefined
That’s right
That's why functions of type
a -> b can't be implemented
(for the rest of eternity, I will elide all sorts of
undefined implementations)
undefineds?
The idea being you can go back and fill them in?
not useful at all, but every time this topic is discussed some Besserwisser points them out
haha
I shouldn't call people jerks, I'll edit it :)
Could you say one useful implementation?
It is basically
f (f (f (f (f ..)))
it should be renamed Hack for SEO reasons
this alone
could change the coding market
for decades
years at the least
...
I am looking for the guidance in hardware software co-design field using Haskell
any papers or tutorials which I can follow to get started?
Hello, can anyone help me out with this?
choose :: PickingMonad m => [a] -> m a
choose = undefined
- choose xs should run without error for any non-empty list xs :: [a] (for the empty list it can do anything)
- in the case of the monad m = IO, choose xs should run in time proportional to the length of xs
- in the case of the monad m = Dist, choose xs :: Dist a should compute a (not necessarily normalised)
distribution where each value in xs is assigned a probability proportional to the number of times it occurs in xs.
That is, prob xs x should be equal to k / n, where k = length [y | y <- xs, x == y] and n = length xs.I have written the code in the way like this, not sure if its correct concept, but i get errors.
choose :: PickingMonad m => [a] -> m a
choose [] = error "Error"
choose xs = do
let n = length xs
i <- getRandomR (0, n-1)
return (xs !! i)