-
Notifications
You must be signed in to change notification settings - Fork 456
add initialization heuristic that samples around best points #987
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
Conversation
|
This pull request was exported from Phabricator. Differential Revision: D32548535 |
…torch#987) Summary: Pull Request resolved: meta-pytorch#987 This adds a heuristic for generating initial conditions by adding gaussian noise to the best points (in addition to the standard sobol initial conditions), which improves optimization performance on some benchmark problems (e.g. ZDT1). This is turned off by default currently. Differential Revision: D32548535 fbshipit-source-id: d63e30f805580f7a25a0d949fea45981ff05ef7c
|
This pull request was exported from Phabricator. Differential Revision: D32548535 |
a0e4c38 to
edd49e4
Compare
Codecov Report
@@ Coverage Diff @@
## main #987 +/- ##
=========================================
Coverage 100.00% 100.00%
=========================================
Files 110 110
Lines 8569 8645 +76
=========================================
+ Hits 8569 8645 +76
Continue to review full report at Codecov.
|
…torch#987) Summary: Pull Request resolved: meta-pytorch#987 This adds a heuristic for generating initial conditions by adding gaussian noise to the best points (in addition to the standard sobol initial conditions), which improves optimization performance on some benchmark problems (e.g. ZDT1). This is turned off by default currently. Differential Revision: D32548535 fbshipit-source-id: 2a753754c89e5e420ef9d422746d9f4ba2477ce5
edd49e4 to
9063738
Compare
|
This pull request was exported from Phabricator. Differential Revision: D32548535 |
…torch#987) Summary: Pull Request resolved: meta-pytorch#987 This adds a heuristic for generating initial conditions by adding gaussian noise to the best points (in addition to the standard sobol initial conditions), which improves optimization performance on some benchmark problems (e.g. ZDT1). This is turned off by default currently. Differential Revision: D32548535 fbshipit-source-id: 2cf894ef0230a92458613f46c6023f248bdbfd28
…torch#987) Summary: Pull Request resolved: meta-pytorch#987 This adds a heuristic for generating initial conditions by adding gaussian noise to the best points (in addition to the standard sobol initial conditions), which improves optimization performance on some benchmark problems (e.g. ZDT1). This is turned off by default currently. Differential Revision: D32548535 fbshipit-source-id: 8f7f970d0e035d9b2277b26a5a90c3d1a745dbb7
|
This pull request was exported from Phabricator. Differential Revision: D32548535 |
9063738 to
94b9dfe
Compare
…torch#987) Summary: Pull Request resolved: meta-pytorch#987 This adds a heuristic for generating initial conditions by adding gaussian noise to the best points (in addition to the standard sobol initial conditions), which improves optimization performance on some benchmark problems (e.g. ZDT1). This is turned off by default currently. Differential Revision: D32548535 fbshipit-source-id: 36d4d0a493dd8ea5c45f73ac6eebb9250a658342
94b9dfe to
6d35ce9
Compare
|
This pull request was exported from Phabricator. Differential Revision: D32548535 |
…torch#987) Summary: Pull Request resolved: meta-pytorch#987 This adds a heuristic for generating initial conditions by adding gaussian noise to the best points (in addition to the standard sobol initial conditions), which improves optimization performance on some benchmark problems (e.g. ZDT1). This is turned off by default currently. Differential Revision: D32548535 fbshipit-source-id: f1df8950c34fa261f95fc3df28aba0085f3740de
…torch#987) Summary: Pull Request resolved: meta-pytorch#987 This adds a heuristic for generating initial conditions by adding gaussian noise to the best points (in addition to the standard sobol initial conditions), which improves optimization performance on some benchmark problems (e.g. ZDT1). This is turned off by default currently. Differential Revision: D32548535 fbshipit-source-id: 5c65a294b227cd719bdd36ba579af8cfb701ea92
…torch#987) Summary: Pull Request resolved: meta-pytorch#987 This adds a heuristic for generating initial conditions by adding gaussian noise to the best points (in addition to the standard sobol initial conditions), which improves optimization performance on some benchmark problems (e.g. ZDT1). This is turned off by default currently. Differential Revision: D32548535 fbshipit-source-id: 903791db2b5ae9d2c3daf880118545bfcac8b4db
6d35ce9 to
b3f3808
Compare
|
This pull request was exported from Phabricator. Differential Revision: D32548535 |
…torch#987) Summary: Pull Request resolved: meta-pytorch#987 This adds a heuristic for generating initial conditions by adding gaussian noise to the best points (in addition to the standard sobol initial conditions), which improves optimization performance on some benchmark problems (e.g. ZDT1). This is turned off by default currently. Differential Revision: D32548535 fbshipit-source-id: a5c3ad5a608fb3ac453e9092d9fc4943a8720a88
…torch#987) Summary: Pull Request resolved: meta-pytorch#987 This adds a heuristic for generating initial conditions by adding gaussian noise to the best points (in addition to the standard sobol initial conditions), which improves optimization performance on some benchmark problems (e.g. ZDT1). This is turned off by default currently. Differential Revision: D32548535 fbshipit-source-id: 43f1f0cb8ddef807ea88c11ef8c3fcf34c3f50f8
…torch#987) Summary: Pull Request resolved: meta-pytorch#987 This adds a heuristic for generating initial conditions by adding gaussian noise to the best points (in addition to the standard sobol initial conditions), which improves optimization performance on some benchmark problems (e.g. ZDT1). This is turned off by default currently. Differential Revision: D32548535 fbshipit-source-id: aaa57aa9327a49c0404f53861b1ddc5599c2149e
…torch#987) Summary: Pull Request resolved: meta-pytorch#987 This adds a heuristic for generating initial conditions by adding gaussian noise to the best points (in addition to the standard sobol initial conditions), which improves optimization performance on some benchmark problems (e.g. ZDT1). This is turned off by default currently. Differential Revision: D32548535 fbshipit-source-id: baffb811be998c4fe9af130ebacfb4a84f16f919
b3f3808 to
5ef906a
Compare
|
This pull request was exported from Phabricator. Differential Revision: D32548535 |
…torch#987) Summary: Pull Request resolved: meta-pytorch#987 This adds a heuristic for generating initial conditions by adding gaussian noise to the best points (in addition to the standard sobol initial conditions), which improves optimization performance on some benchmark problems (e.g. ZDT1). This is turned off by default currently. Differential Revision: D32548535 fbshipit-source-id: 2a3db4ca05b74bec2d4758d7b2eb2de33671dc20
…torch#987) Summary: Pull Request resolved: meta-pytorch#987 This adds a heuristic for generating initial conditions by adding gaussian noise to the best points (in addition to the standard sobol initial conditions), which improves optimization performance on some benchmark problems (e.g. ZDT1). This is turned off by default currently. Reviewed By: Balandat Differential Revision: D32548535 fbshipit-source-id: 3c39c7638f789bd6e64f69c81209ade0169a9e6f
|
This pull request was exported from Phabricator. Differential Revision: D32548535 |
5ef906a to
1b5b67f
Compare
…torch#987) Summary: Pull Request resolved: meta-pytorch#987 This adds a heuristic for generating initial conditions by adding gaussian noise to the best points (in addition to the standard sobol initial conditions), which improves optimization performance on some benchmark problems (e.g. ZDT1). This is turned off by default currently. Reviewed By: Balandat Differential Revision: D32548535 fbshipit-source-id: 6de907a6523e1222c665b34ef01913763e4d43fb
|
This pull request was exported from Phabricator. Differential Revision: D32548535 |
1b5b67f to
afe040d
Compare
…torch#987) Summary: Pull Request resolved: meta-pytorch#987 This adds a heuristic for generating initial conditions by adding gaussian noise to the best points (in addition to the standard sobol initial conditions), which improves optimization performance on some benchmark problems (e.g. ZDT1). This is turned off by default currently. Differential Revision: D32548535 fbshipit-source-id: c629f0eb5e87a5e675faa3a9471f1044421c6964
…torch#987) Summary: Pull Request resolved: meta-pytorch#987 This adds a heuristic for generating initial conditions by adding gaussian noise to the best points (in addition to the standard sobol initial conditions), which improves optimization performance on some benchmark problems (e.g. ZDT1). This is turned off by default currently. Differential Revision: D32548535 fbshipit-source-id: d49648e7dbc9d515fc590172e284519f8018adff
…torch#987) Summary: Pull Request resolved: meta-pytorch#987 This adds a heuristic for generating initial conditions by adding gaussian noise to the best points (in addition to the standard sobol initial conditions), which improves optimization performance on some benchmark problems (e.g. ZDT1). This is turned off by default currently. Differential Revision: D32548535 fbshipit-source-id: 27c9d8c696f1a931f796021ac640ca3a234d8f7b
…torch#987) Summary: Pull Request resolved: meta-pytorch#987 This adds a heuristic for generating initial conditions by adding gaussian noise to the best points (in addition to the standard sobol initial conditions), which improves optimization performance on some benchmark problems (e.g. ZDT1). This is turned off by default currently. Reviewed By: Balandat Differential Revision: D32548535 fbshipit-source-id: 548706bc23cd0184708af888d65206de6b23c23b
Summary:
This adds a heuristic for generating initial conditions by adding gaussian noise to the best points (in addition to the standard sobol initial conditions), which improves optimization performance on some benchmark problems (e.g. ZDT1).
This is turned off by default currently.
Differential Revision: D32548535