| 1 |
Files in this directory demonstrate basic use of mystic. The most common |
| 2 |
case demonstrated is fitting a standard test function from mystic.models, |
| 3 |
however the use of constraints, the use of ensemble solvers, and the use |
| 4 |
of parallel computing is also demonstrated. |
| 5 |
|
| 6 |
|
| 7 |
== Notes on mystic examples == |
| 8 |
Dependencies: |
| 9 |
- Several of the examples require matplotlib to be installed. |
| 10 |
- For the examples that use matplotlib, see trac ticket #36 for more details. |
| 11 |
|
| 12 |
Other dependencies: |
| 13 |
- Examples with prefix "example" are part of the tutorial (TRY THESE FIRST). |
| 14 |
- All examples with prefix "example" should run without new dependencies. |
| 15 |
- All examples with prefix "test_" should run without new dependencies. |
| 16 |
- All examples with prefix "gplot_" requre gnuplot-py for visualization. |
| 17 |
|
| 18 |
Exceptions to the rule: |
| 19 |
- The following examples also require scipy to be installed: |
| 20 |
. test_lorentzian.py |
| 21 |
. test_mogi_anneal.py, |
| 22 |
|
| 23 |
Special examples: |
| 24 |
- All examples with prefix "rosetta_" require park to be installed. (tests on version park-1.2). Run with "--park" to execute with park. See "--help" for more options. |
| 25 |
|
| 26 |
|
| 27 |
------------------------------------------------------------------------------- |
| 28 |
|
| 29 |
Notes on the "ffit" tests/examples: |
| 30 |
- test_ffit: The fitting problem whose exact solution is 8th order Chebyshev polynomial of the first kind. This example uses a Ctrl-C signal handler. Try ctrl-c as the differential_evolution strategy is running. |
| 31 |
- test_ffit2: The fitting problem whose exact solution is 16th order Chebyshev polynomial of the first kind. Also uses the signal_handler. |
| 32 |
- test_ffitB: Same as test_ffit.py, but uses DifferentialEvolutionSolver2 instead of DifferentialEvolutionSolver. |
| 33 |
- test_ffitC: Same as test_ffit.py, but uses scipy_optimize.fmin. |
| 34 |
- test_ffitD: Same as test_ffit.py, but uses scipy_optimize.diffev. |
| 35 |
|
| 36 |
|
| 37 |
Notes on the "mogi" tests/examples: |
| 38 |
- test_mogi.py: One mogi source with noise, comparison between DE and Conjugate Gradient, Simplex, and least squares (Levenberg Marquardt). CG / lsq don't work very well. lsq should work when bounds on the parameters are given, but minpack (wrapped by scipy) version doesn't seem to support bounds. |
| 39 |
- test_mogi_anneal.py: tests with scipy simulated annealing, but hasn't been tuned, so again, doesn't work at all. |
| 40 |
- test_mogi2.py: two mogi sources |
| 41 |
- test_mogi3.py: reimplements test_mogi |
| 42 |
|
| 43 |
|
| 44 |
# end of file |