Release that supports Julia 0.6 #14
Conversation
abhijithanilkumar
commented
May 21, 2017
•
abhijithanilkumar
commented
- Use StaticArrays instead of FixedSizeArrays
- Fix Deprecation warnings
- Fix type issues
test/runtests.jl
Outdated
| @@ -151,8 +151,8 @@ jagmesh_adj = jagmesh() | |||
| for i in 1:size(a,1) | |||
| p = Int32[] | |||
| for e in collect(edges(g)) | |||
| if e[1] == i | |||
| push!(p,e[2]) | |||
| if e.src == i | |||
There was a problem hiding this comment.
This should be src(e) for LightGraphs - use the accessor function, not the internal field.
Same with L155 - dst(e).
Codecov Report
@@ Coverage Diff @@
## master #14 +/- ##
=========================================
+ Coverage 97.22% 97.53% +0.3%
=========================================
Files 7 7
Lines 324 324
=========================================
+ Hits 315 316 +1
+ Misses 9 8 -1
Continue to review full report at Codecov.
|
|
Changes Unknown when pulling e563502 on v0.6-dev into ** on master**. |
|
Changes Unknown when pulling e47a426 on v0.6-dev into ** on master**. |
1 similar comment
|
Changes Unknown when pulling e47a426 on v0.6-dev into ** on master**. |
src/stress.jl
Outdated
| map(D) do d | ||
| x = T(d^(-2.0)) | ||
| isfinite(x) ? x : zero(T) | ||
| x = Float64(d^(-2.0)) |
There was a problem hiding this comment.
why would you want to restrict it only to Float64?
There was a problem hiding this comment.
When generalizing, in 0.6 it's creating a sparse matrix of type {Any,Int64}. In 0.5 it was of type {T,Int64}. So in 0.6 this is throwing an error while calling zeros() function later in weightedlaplacian. Also, isn't it safe to assume that the output of d^-2.0 is always floating point?
There was a problem hiding this comment.
Also, isn't it safe to assume that the output of d^-2.0 is always floating point?
Float64 is not the only float type in julia...
And Any as the matrix eltype shouldn't happen, so you're just working around a bug that must have happened earlier!
There was a problem hiding this comment.
Not the eltype, map function is generating different outputs in Julia 0.5 and 0.6
In 0.5, it generates sparse matrix with Float elements :
julia> function fx(a,t=eltype(a))
map(a) do d
x=t(d^(-2.0))
isfinite(x)?x:zero(t)
end
end
fx (generic function with 2 methods)
julia> fx(a,Float32)
30×30 sparse matrix with 116 Float32 nonzero entries:
[2 , 1] = 1.0
[3 , 1] = 1.0
[4 , 1] = 1.0
[5 , 1] = 1.0
[6 , 1] = 1.0
[7 , 1] = 1.0
[8 , 1] = 1.0
In 0.6 it is generating matrix with Any elements :
julia> function fx(a,t=eltype(a))
map(a) do d
x=t(d^(-2.0))
isfinite(x)?x:zero(t)
end
end
fx (generic function with 2 methods)
julia> fx(a,Float32)
30×30 SparseMatrixCSC{Any,Int64} with 116 stored entries:
[2 , 1] = 1.0
[3 , 1] = 1.0
[4 , 1] = 1.0
[5 , 1] = 1.0
[6 , 1] = 1.0
[7 , 1] = 1.0
[8 , 1] = 1.0
What do you suggest?
There was a problem hiding this comment.
It must be d^t(-2.0) then...
There was a problem hiding this comment.
Tried the above suggestions and still no luck :/
There was a problem hiding this comment.
Sorry, I have absolutely no time to care about this package right now.
If you can't figure it out, I guess that's then just how it is.
There was a problem hiding this comment.
Thanks for your time :) I'll try to find a way.
There was a problem hiding this comment.
I think I've figured out how to do this:
julia> function fx(a,t=eltype(a))::SparseMatrixCSC{t, Int64}
map(a) do d
x=t(d^(-2.0))
isfinite(x)?x:zero(t)
end
end
fx (generic function with 2 methods)
julia> g = Graph(3,3)
{3, 3} undirected simple Int64 graph
julia> fx(adjacency_matrix(g), Float32)
3×3 SparseMatrixCSC{Float32,Int64} with 6 stored entries:
[2, 1] = 1.0
[3, 1] = 1.0
[1, 2] = 1.0
[3, 2] = 1.0
[1, 3] = 1.0
[2, 3] = 1.0
Return type annotation.
There was a problem hiding this comment.
Thank you for the suggestion! That fixes the issue.
|
Changes Unknown when pulling 7474d77 on v0.6-dev into ** on master**. |
