Skip to content

Commit ba87a50

Browse files
Balandatfacebook-github-bot
authored andcommitted
Changelog for v0.3.3 (#622)
Summary: Pull Request resolved: #622 Reviewed By: eytan Differential Revision: D25390481 Pulled By: Balandat fbshipit-source-id: adb1b1f7e2992f078f22393ab25a8d686d5e2980
1 parent 9edf728 commit ba87a50

File tree

1 file changed

+39
-0
lines changed

1 file changed

+39
-0
lines changed

CHANGELOG.md

Lines changed: 39 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -2,6 +2,45 @@
22

33
The release log for BoTorch.
44

5+
6+
## [0.3.3] - Dec 8, 2020
7+
8+
Contextual Bayesian Optimization, Input Warping, TuRBO, sampling from polytopes.
9+
10+
#### Compatibility
11+
* Require PyTorch >=1.7 (#614).
12+
* Require GPyTorch >=1.3 (#614).
13+
14+
#### New Features
15+
* Models (LCE-A, LCE-M and SAC ) for Contextual Bayesian Optimziation (#581).
16+
* Implements core models from:
17+
[High-Dimensional Contextual Policy Search with Unknown Context Rewards using Bayesian Optimization](https://proceedings.neurips.cc/paper/2020/hash/faff959d885ec0ecf70741a846c34d1d-Abstract.html).
18+
Q. Feng, B. Letham, H. Mao, E. Bakshy. NeurIPS 2020.
19+
* See Ax for usage of these models.
20+
* Hit and run sampler for uniform sampling from a polytope (#592).
21+
* Input warping:
22+
* Core functionality (#607).
23+
* Kumaraswamy Distribution (#606).
24+
* Tutorial (8f34871652042219c57b799669a679aab5eed7e3).
25+
* TuRBO-1 tutorial (#598).
26+
* Implements the method from [Scalable Global Optimization via
27+
Local Bayesian Optimization](https://proceedings.neurips.cc/paper/2019/file/6c990b7aca7bc7058f5e98ea909e924b-Paper.pdf).
28+
D. Eriksson, M. Pearce, J. Gardner, R. D. Turner, M. Poloczek. NeurIPS 2019.
29+
30+
#### Bug fixes
31+
* Fix bounds of `HolderTable` synthetic function (#596).
32+
* Fix `device` issue in MOO tutorial (#621).
33+
34+
#### Other changes
35+
* Add `train_inputs` option to `qMaxValueEntropy` (#593).
36+
* Enable gpytorch settings to override BoTorch defaults for `fast_pred_var` and `debug` (#595).
37+
* Rename `set_train_data_transform` -> `preprocess_transform` (#575).
38+
* Modify `_expand_bounds()` shape checks to work with >2-dim bounds (#604).
39+
* Add `batch_shape` property to models (#588).
40+
* Modify `qMultiFidelityKnowledgeGradient.evaluate()` to work with `project`, `expand` and `cost_aware_utility` (#594).
41+
* Add list of papers using BoTorch to website docs (#617).
42+
43+
544
## [0.3.2] - Oct 23, 2020
645

746
Maintenance Release

0 commit comments

Comments
 (0)