Denis Rosset

They are now (or in the near future) commented out under "TODO"s, at least for FixedPoint and Real. We should revisit, document and verify their implementation.

#334 Big architectural change proposed by @tixxit in 2014 #396 revival by @pleira in 2015 #569 Support of complex vector spaces #658 Normed vector space should have two scalar parameters...

hashCode and spire's Eq

2

I have a use case where I need to group elements from a sequence (think `groupBy`) in a group theory library I'm writing, that depends on Spire. I can think...

Two nice ideas to improve test coverage: #627 Use homomorphisms between algebraic structures to provide additional tests #642 Have tests for limited range primitives

Future plans for Interval

2

Now that I know this part by heart, what could be the next steps ? I'm thinking of: - [ ] add ref. to literature about implementation of open/closed intervals...

These two lines are equivalent in `object Polynomial`: ```scala def linear[@sp(Double) C: Eq: Semiring: ClassTag](c1: C, c0: C): Polynomial[C] = Polynomial(Map((1, c1), (0, c0))) def quadratic[@sp(Double) C: Eq: Semiring: ClassTag](c1:...

We had the following special case: ```scala final override def gcd(x: Polynomial[C], y: Polynomial[C])(implicit ev: Eq[Polynomial[C]]): Polynomial[C] = { val result = EuclideanRing.euclid(x, y)(ev, self) if (result.degree > 0) {...

What is exactly `Jet`?

1

The status of `Jet` is unclear. There are conflicting indications that: 1. `Jet` is intended to be a substitute of a real number in computations, so that derivatives can be...

Turn inlining on only in production builds

1

as per Scala 2.12 release notes: > We recommend enabling inlining only in production builds, as sbt’s incremental compilation does not track dependencies introduced by inlining.

I'd like to formalize the notion of complex conjugation in Spire. Most of the things below are already implemented in the `Complex` data type. I propose to implement some parts...