Ticket #4827 (closed enhancement: fixed)
[with patch, positive review] add L-BFGS-B bound constraint solver to minimize_constraint
| Reported by: | schilly | Owned by: | jkantor |
|---|---|---|---|
| Priority: | minor | Milestone: | sage-3.3 |
| Component: | numerical | Keywords: | |
| Cc: | Work issues: | ||
| Report Upstream: | Reviewers: | ||
| Authors: | Merged in: | ||
| Dependencies: | Stopgaps: |
Description
adding scipy's l-bfgs-b large scale bound constraint solver, small change in docstring: bounds are better off in tuples.
Attachments
-
lbfgsb-v1.patch
(3.7 KB) -
added by schilly 4 years ago.
- adds lbfgsb to minimize_constraint
-
trac_4827-optimize-doc.patch
(1.3 KB) -
added by jason 4 years ago.
- apply on top of previous patch
Change History
Changed 4 years ago by schilly
-
attachment
lbfgsb-v1.patch
added
Changed 4 years ago by jason
-
attachment
trac_4827-optimize-doc.patch
added
apply on top of previous patch
comment:1 follow-up: ↓ 2 Changed 4 years ago by jason
- Summary changed from [with patch, needs review] add L-BFGS-B bound constraint solver to minimize_constraint to [with patch, positive review] add L-BFGS-B bound constraint solver to minimize_constraint
Looks good to me.
For another ticket: the minimize_constrained arguments should have x0 before cons to parallel the argument structure of the other minimze functions. Also, we should maybe look into using openopt at some point since the syntax is then unified.
The small doc patch I attached makes the documentation of the function arguments in the same order as the function arguments. I don't think it needs to be reviewed.
comment:2 in reply to: ↑ 1 Changed 4 years ago by schilly
Replying to jason:
Also, we should maybe look into using openopt at some point since the syntax is then unified.
yeahr, tell me if somebody is working on that or you need help. maybe i can look into it. probably the most difficult part is to detect/register optional solvers inside sage, or talking to installed ones...
The small doc patch I attached makes the documentation of the function arguments in the same order as the function arguments. I don't think it needs to be reviewed.
well, fwiw, +1 from me ;)
adds lbfgsb to minimize_constraint