1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
|
Files in this directory demonstrate use cases for some of mystic's "advanced"
features, such as using dimensional collapse detectors, building a surrogate
for an unknown surface, or using support vectors in optimization.
== Notes on mystic examples3 ==
Naming convention:
- `3D` denotes use of 3-D toy model in surrogate.py
- `5D` denotes use of 5-D toy model in toys.py
- `GP` denotes use of sklearn GP regression instead of RBF interpolation
- `NN` denotes use of sklearn MLP regression instead of RBF interpolation
- 'ML' denotes use of iterative surrogate-assisted optimization
- `c0mean` denotes a (backpropagation) calculation of mean value of an input
- `collapse` denotes use of dimensional collapse detectors
- `mean` denotes a calculation of mean value of an output
- `ouq` denotes use of high-level UQ interface to find bounds, surrogates
- `pandas` denotes mystic extensions/interface to pandas
- `pof` denotes a probability of failure calculation
- `searcher` denotes optimizer-driven discovery of all extrema [DEPRECATED]
- `sklearn` denotes mystic extensions/interface to sklearn
- `surface` denotes optimizer-driven surface interpolation [DEPRECATED]
- `svc` denotes use of support vectors in classification
- `svr` denotes use of support vectors in regression
- `sparsity` denotes use of the sparsity sampler
- `xrd` denotes the lattice parameter toy model
OUQ calculations:
- test_*_ub_*: find upper bound on (mean of input/output, or PoF) [examples5]
- test_error*: find model error
- test_expect: find expected value
- test_expected*: find bound on expected value
- test_expected_error* find bound on expected error
- test_glb*: find greatest lower bound
- test_gub*: find greatest upper bound
- test_llb*: find least lower bound
- test_lub*: find least upper bound
- test_pof: find probability of failure (PoF)
- xrd_design*: find average and bounds on expected value and bounds
Miscellaneous:
- test_cache: demonstrate use and customization of mystic.cache
- test_improve_score: workflow to retry/update sklearn estimators
- xrd_opt*: find minimum using ensemble optimization
- xrd_optML*: find minimum using iterative surrogate improvement
- xrd_*_db: inspect expected value databases generated with xrd_design*
External dependencies:
- examples with "svc", "svr", or "sparsity" in the name require `matplotlib`.
- examples with "surface" and "searcher" require `scipy` and `matplotlib`.
- examples with "pandas" require `pandas` and `matplotlib`.
- otherwise, assume `sklearn` and `matplotlib` are required.
In-place dependencies for ouq* calculations:
- dataset: extensions on dataset operations
- emulators: toy models emulating XRD calculations
- estimator: high-level OUQ estimator (requires sklearn)
- interpolator: high-level OUQ interpolator
- misc: miscellaneous constraints and definitions for OUQ calculation
- ml: lower-level interface to sklearn, used by estimator
- noisy: add noise for OUQ noisy model
- plotter: high-level interface to plotting learned surrogates
- spec*: same as misc, used in test_*_ub_* OUQ calculations
- surrogate: hypervelocity impact model, used in test_*_ub_* OUQ calculations
- toys: simple toy models for OUQ calculations
|
