-
Notifications
You must be signed in to change notification settings - Fork 231
Do not (mis)use objective as state #1212
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Do not (mis)use objective as state #1212
Conversation
935c749 to
93e8f31
Compare
| end | ||
|
|
||
| if g_calls(d) > 0 && !all(isfinite, gradient(d)) | ||
| if hasproperty(state, :g_x) && !all(isfinite, state.g_x) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
could in principle add methods to isfinite_gradient and isfinite_hessian that you introduced for the objective? I suppose they are sufficiently similar in nature to share the function?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
if it's only used once of course... then it's maybe overkill :) I suppose in termination code as well..
c4d1eec to
d2071a1
Compare
8c40315 to
60a5fef
Compare
Benchmark Results (Julia vlts)Time benchmarks
Memory benchmarks
A plot of the benchmark results has been uploaded as an artifact at . |
938f2ec to
95ca2d8
Compare
Codecov Report❌ Patch coverage is Additional details and impacted files@@ Coverage Diff @@
## master #1212 +/- ##
==========================================
- Coverage 87.32% 86.77% -0.55%
==========================================
Files 45 45
Lines 3558 3539 -19
==========================================
- Hits 3107 3071 -36
- Misses 451 468 +17 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
95ca2d8 to
92ae7aa
Compare
92ae7aa to
6182632
Compare
|
I suppose the artifacts need to be uniquely named |
|
Indeed, I already opened a PR on Friday: MilesCranmer/AirspeedVelocity.jl#120 |
Very much WIP, and I might have to start over again. Fixes #1210 (comment) by not (mis)using objective functions as states. Instead, this PR proposes to add
f_xandg_xfields to the optimization states (where needed), similar to already existingx_previous,f_x_previousandg_x_previousfields.