Skip to content

Releases: pcarruscag/FADO

FADO version 1.2.5

12 Feb 02:28
ee0eb0b
Compare
Choose a tag to compare

Features

  • Monitor the value of functions that do not participate actively in the optimization Driver.addMonitor

FADO version 1.2.4

08 Dec 16:09
d4dfa56
Compare
Choose a tag to compare

Features

  • Allow a user-defined destination file name in ExternalRun.addData

FADO version 1.2.3

06 May 18:12
Compare
Choose a tag to compare

Maintenance

  • Update example for SU2 version > 7.1.0;
  • Add an example for shape optimization of an FSI problem.

FADO version 1.2.2

20 Nov 09:45
Compare
Choose a tag to compare

Bug fixes

  • Fix dangling evaluations when using concurrent execution (if one execution failed early).
  • Ensure that the default function value is returned if it fails during concurrent execution (for soft failure mode only).

FADO version 1.2.1

17 Jul 20:58
Compare
Choose a tag to compare

Features

  • Class LabeledTableReader to read data from csv-like files based on column headers;
  • Generalize user-defined pre/post driver actions (driver.setUserPreProcessFun) to accept both strings (shell commands) and callable objects.

Maintenance

  • Consistent use of comparison operators (is and ==);
  • Collect all examples under one common directory.

FADO version 1.2.0

22 May 14:31
Compare
Choose a tag to compare

Features

  • Driver class (ScipyDriver) for easy interfacing with SciPy optimizers, especially SLSQP;
  • Python docstrings on all important classes and functions, plus an utility function (printDocumentation) to automatically display the documentation, e.g. >>> printDocumentation(ScipyDriver).

Bug fixes

  • Simultaneous evaluation of gradients was not working for the IpoptDriver.

FADO version 1.1.0

21 Apr 21:19
Compare
Choose a tag to compare

Features

  • Driver class (IpoptDriver) for easy interfacing with Ipopt (via IPyOpt);
  • New file parser (ArrayLabelReplacer), replaces a label by an array of numeric values;
  • Adds "soft-failure" functionality (driver.setFailureMode("SOFT")), instead of raising an exception when a function evaluation fails, the code uses a fallback value (function.setDefaultValue(x)) and the previous value of the gradient.

Bug fixes

  • The re-try mechanism (that re-runs an evaluation should it fail) was not working.

FADO version 1.0.0

22 Feb 11:26
Compare
Choose a tag to compare
v1.0.0

add more info to readme