Fix QR based fitting #559
Fix QR based fitting #559
Conversation
andreasnoack
force-pushed
the
an/norank
branch
2 times, most recently
from
33d1584
to
35c3e94
Compare
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## master #559 +/- ##
==========================================
- Coverage 90.03% 89.95% -0.09%
==========================================
Files 8 8
Lines 1124 1105 -19
==========================================
- Hits 1012 994 -18
+ Misses 112 111 -1 ☔ View full report in Codecov by Sentry. |
andreasnoack
requested review from
devmotion
and removed request for
mousum-github
| fill!(p.delbeta, 0) | ||
| p.delbeta[1:rnk] = R \ view(p.qr.Q'r, 1:rnk) |
There was a problem hiding this comment.
Maybe one could save a few allocations here?
| fill!(p.delbeta, 0) | |
| p.delbeta[1:rnk] = R \ view(p.qr.Q'r, 1:rnk) | |
| mul!(view(p.delbeta, 1:rnk), view(p.qr.Q, :, 1:rnk)', r) | |
| ldiv!(R, view(p.delbeta, 1:rnk)) | |
| fill!(@view(p.delbeta[(rnk + 1):end]), 0) |
There was a problem hiding this comment.
I think a view of a Householder matrix would end up hitting a fallback multiplicatino which would be very costly.
src/linpred.jl
Outdated
| end | ||
| p.delbeta[1:rnk] = R \ (p.qr.Q'*r̃)[1:rnk] |
There was a problem hiding this comment.
Maybe
| p.delbeta[1:rnk] = R \ (p.qr.Q'*r̃)[1:rnk] | |
| mul!(view(p.delbeta, 1:rnk), view(p.qr.Q, :, 1:rnk)', r̃) | |
| ldiv!(R, view(p.delbeta, 1:rnk)) |
?
There was a problem hiding this comment.
We can't make a view into Q, see above, but we can make a view into the result.
Avoid erroring out for low rank design matrices when dropcollinear=false. Avoid unnecessary triangular solves. Avoid indexing in the Q. Avoid slicing R matrix in a way that triggers a minimum norm solution. Remove unnecessary temporaries.
| p.delbeta = Rinv * Rinv' * p.delbeta | ||
| ỹ = sqrtW * r | ||
| p.qr = qr!(p.scratchm1) | ||
| p.delbeta = p.qr \ ỹ |
There was a problem hiding this comment.
It's not a square problem so \ doing a bit more than just ldiv! and a copy, see https://github.com/JuliaLang/julia/blob/2a2878c143b87e5184565c895d090aab6e9017e9/stdlib/LinearAlgebra/src/LinearAlgebra.jl#L704-L731
| fill!(p.delbeta, 0) | ||
| p.delbeta[1:rnk] = R \ view(p.qr.Q'r, 1:rnk) |
There was a problem hiding this comment.
I think a view of a Householder matrix would end up hitting a fallback multiplicatino which would be very costly.
src/linpred.jl
Outdated
| end | ||
| p.delbeta[1:rnk] = R \ (p.qr.Q'*r̃)[1:rnk] |
There was a problem hiding this comment.
We can't make a view into Q, see above, but we can make a view into the result.
andreasnoack
force-pushed
the
an/norank
branch
from
9bbc134
to
fffd07d
Compare
andreasnoack
force-pushed
the
an/norank
branch
2 times, most recently
from
dcc9e21
to
37249f5
Compare
andreasnoack
force-pushed
the
an/norank
branch
from
37249f5
to
2b61e90
Compare
|
@devmotion I finally found a way to download the test dataset reliably so this one should be good to go. |
Avoid erroring out for low rank design matrices when dropcollinear=false.
Avoid unnecessary triangular solves. Avoid indexing in the Q.
Avoid slicing R matrix in a way that triggers a minimum norm solution.
Remove unnecessary temporaries.
Fixes #558
I removed the white space so please review with https://github.com/JuliaStats/GLM.jl/pull/559/files?w=1