Groebner.jl integration #5
Conversation
Codecov Report
@@ Coverage Diff @@
## master #5 +/- ##
=========================================
Coverage ? 75.05%
=========================================
Files ? 7
Lines ? 938
Branches ? 0
=========================================
Hits ? 704
Misses ? 234
Partials ? 0 Continue to review full report at Codecov.
|
| MPoly{T}(x::CoeffType) where {T} = MPoly(eltype(T)(x)) | ||
| terms(x::MPoly) = x.terms | ||
| function nvariables(p::MPoly) | ||
| if monomialtype(p) <: PackedMonomial |
There was a problem hiding this comment.
Why only support PackedMonomial here?
| PackedMonomial{L,E}(bits::NTuple{K,UInt64}) where {L,E,K} = new{L,new_E(Val(E)),K}(bits) | ||
| PackedMonomial{L,E,K}(bits::NTuple{K,UInt64}) where {L,E,K} = new{L,new_E(Val(E)),K}(bits) | ||
| end | ||
| nvariables(x::PackedMonomial{L}) where L = L |
There was a problem hiding this comment.
This definition is not the same as the nvariables definition for Monomial.
L is the maximum number of supported variables.
There was a problem hiding this comment.
Yeah, I just added those quickly so I can use Groebner.jl :p it's not optimized.
There was a problem hiding this comment.
Also, computing nvariables for polynomial can be expensive. We have to take the union.
|
Mind adding tests and adding an example? |
|
After loading SIMDPolynomials and Groebner julia> x1, x2 = [PackedMonomial{2,7}(i) for i in 0:1];
julia> system = [10*x1*x2^2 - 11*x1 + 10,
10*x1^2*x2 - 11*x2 + 10];
julia> groebner(system)
3-element Vector{MPoly{Vector{SIMDPolynomials.Term{Int64, PackedMonomial{2, 7, 1}}}}}:
11x₀² - 11x₁² - 10x₀ + 10x₁
110x₁³ + 100x₀x₁ - 100x₁² - 121x₁ + 110
10x₀x₁² - 11x₀ + 10 |
Co-authored-by: Chris Elrod <elrodc@gmail.com>
No description provided.