#2853 closed defect (fixed)
[with patch, positive review] A correction to the weight of crystal elements for type A and a speedup for all types
Reported by: | bump | Owned by: | mhansen |
---|---|---|---|
Priority: | minor | Milestone: | sage-3.0 |
Component: | combinatorics | Keywords: | |
Cc: | sage-combinat | Merged in: | |
Authors: | Reviewers: | ||
Report Upstream: | Work issues: | ||
Branch: | Commit: | ||
Dependencies: | Stopgaps: |
Description
For Cartan Types 'A' a problem with the weight function of crystals was described here:
http://groups.google.com/group/sage-combinat-devel/browse_thread/thread/7cdfe075257ba963?hl=en
The method of correcting this problem was to hard-code the weight in the crystals of letters, and to have the crystals of tensors get the weight of a tensor element by summing the weights of its constituents. This alters the weight for Type A (correcting the defect) and returns the same weight as the old algorithm for other Cartan types.
When the patch was implemented it was found to be 2-3 times faster than the old algorithm.
Attachments (3)
Change History (12)
Changed 13 years ago by
comment:1 Changed 13 years ago by
- Summary changed from A correction to the weight of crystal elements for type A and a speedup for all types to [with patch, needs review] A correction to the weight of crystal elements for type A and a speedup for all types
Changed 13 years ago by
comment:2 Changed 13 years ago by
- Summary changed from [with patch, needs review] A correction to the weight of crystal elements for type A and a speedup for all types to [with patch, needs minor work] A correction to the weight of crystal elements for type A and a speedup for all types
Hi Dan,
I've updated a patch to move the tests for .weight() to the docstring for the definition of weight. There is an issue with ['G',2] though:
Before the patch:
sage: C = CrystalOfLetters(['G',2]) sage: hwvs = C.highest_weight_vectors() sage: wlr = C.weight_lattice_realization() sage: v = hwvs[0]; v 1 sage: v.weight() (1, 0, -1) sage: wlr.weyl_dimension(v.weight()) 7
After the patch:
sage: C = CrystalOfLetters(['G',2]) sage: wlr = C.weight_lattice_realization() sage: hwvs = C.highest_weight_vectors() sage: hwvs [1] sage: v = hwvs[0]; v 1 sage: v.weight() (-1, 0, 1) sage: wlr.weyl_dimension(v.weight()) 0
comment:3 follow-up: ↓ 4 Changed 13 years ago by
I was unable to confirm the problem with G2. I applied the patch in the modified form that you posted, and sage: wlr.weyl_dimension(v.weight()) returns 7 for me.
Dan
comment:4 in reply to: ↑ 3 ; follow-up: ↓ 5 Changed 13 years ago by
Replying to bump:
I was unable to confirm the problem with G2. I applied the patch in the modified form that you posted, and sage: wlr.weyl_dimension(v.weight()) returns 7 for me.
Dan
Hi Dan,
I can confirm that with 3.0.alpha2 and your patch applied wlr.weyl_dimension(v.weight()) returns 0. With what build did you try? Are there any other patches you have in your tree?
Cheers,
Michael
comment:5 in reply to: ↑ 4 Changed 13 years ago by
Hi Dan,
I can confirm that with 3.0.alpha2 and your patch applied wlr.weyl_dimension(v.weight()) returns 0. With what build did you try? Are there any other patches you have in your tree?
Cheers,
Michael
I did this with 3.0-alph0 and no other patches. I can install 3.0.alpha2 and debug the patch.
Dan
Changed 13 years ago by
comment:6 Changed 13 years ago by
It turns out that I had forgotten the G2 patch that went in on Saturday.
I corrected the patch and added it as 2853.1.patch. I believe it is now correct.
Dan
comment:7 Changed 13 years ago by
- Summary changed from [with patch, needs minor work] A correction to the weight of crystal elements for type A and a speedup for all types to [with patch, positive review] A correction to the weight of crystal elements for type A and a speedup for all types
Looks good to me. Apply just 2853.1.patch.
comment:8 Changed 13 years ago by
- Resolution set to fixed
- Status changed from new to closed
Merged in Sage 3.0.alpha3.
comment:9 Changed 12 years ago by
- Cc sage-combinat added
patch correcting problem with weights for crystal graphs of type A and speeding up all Cartan types.