Skip to content

v0.1.17
Compare
Choose a tag to compare
@xingyousong xingyousong released this 11 Jul 18:23
· 49 commits to main since this release
v0.1.17
41c00cc

What's Changed

  • Add decorator @seed_with_default by @copybara-service in #1086
  • Fix bug on allowed version name by @copybara-service in #1092
  • Use TEST_PYPI secrets instead by @copybara-service in #1093
  • Upgrade jax versions to avoid DeviceArray import issues. by @copybara-service in #1094
  • Support for infeasible trials in ensemble_designer by @copybara-service in #1090
  • ...added validation code to satisfy PyType Checks... by @copybara-service in #1096
  • Adds the pe_overwrite_probability parameter to the UCBPEConfig by @copybara-service in #1091
  • Updates the default value for pe_overwrite_probability by @copybara-service in #1101
  • Add HashingInfeasibleExperimenter and ParameterRegionInfeasibleExperimenter to simulate infeasibility behaviors. by @copybara-service in #1104
  • Add to public import infeasible experimenters by @copybara-service in #1105
  • Parameterized the penalty factor and rename it. by @copybara-service in #1103
  • Default trial seeding, without using scalers. by @copybara-service in #1088
  • Implements a variant of GP_UCB_PE that performs exploration via set-optimization by @copybara-service in #1110
  • Supports multi-batch dimensions in the trust region acquisition function. by @copybara-service in #1114
  • Appends the first suggestion to active trials before generating the remaining suggestions in a batch by @copybara-service in #1116
  • Passes prior features to set-acquisition optimization by @copybara-service in #1117
  • Add a wrapper to allow fairer comparisons between Vizier and other raytune algorithms. by @copybara-service in #1118
  • Small update by @copybara-service in #1106
  • Add simple4d convergence test for gp bandit. by @copybara-service in #958
  • Add BenchmarkRecord summarization tools. by @copybara-service in #1120
  • Refactor experimenter factories by @copybara-service in #1121
  • Add xlabel to PlotElement. by @copybara-service in #1123
  • Deprecate predictive_fns.py from Vizier JAX. by @copybara-service in #1126
  • Fix accidental bug in MultiObjectiveNumpyExperimenter by @copybara-service in #1129
  • Aesthetic + small bug fixes to Vizier service by @copybara-service in #1130
  • Revert Twine=5.0.0 to unblock deployment failures: pypa/twine#1125 by @copybara-service in #1131
  • Enables GP_UCB_PE to explore more aggressively in high-noise situations. by @copybara-service in #1133
  • Change k = num_objectives - 1 by @copybara-service in #1134
  • Use reference point and multiple scalarizations in ScalarizedAcquisition. by @copybara-service in #1135
  • Vectorize the scalarization method instead of producing multiple class instances. This should run faster + reduce the jitted graph size (prevent RAM blowups) by @copybara-service in #1136
  • Use 1000 scalarizations as default. Vectorization made it much faster to pass unit tests by @copybara-service in #1138
  • Add Sample acquisition function to GP bandit. by @copybara-service in #1141
  • Linear kernel refactor: Merge VizierGaussianProcess with VizierLinearGaussianProcess by @copybara-service in #1142
  • Add max_scalarized to ScalarizedAcquisition. by @copybara-service in #1143
  • Refactor multiobjective setup by @copybara-service in #1144
  • Prepare for 0.1.7 release by cleaning up multiobjective. by @copybara-service in #1145

Full Changelog: v0.1.16...v0.1.17

Assets 2
Loading