hmer - v1.6.0

News

A technical paper detailing the functionality of the package is now in JSS! See here to read it and if using hmer in an application, please cite this paper.

Bug Fixes

• Behaviour of stepwise addition modified to avoid overfitting in edge cases; the add = TRUE argument can be passed to emulator_from_data to force stepwise addition when creating the regression surface.
• Fixed a visualisation issue in hit_by_wave when the number of waves of emulation exceeds the number of targets.

Minor Changes

• Updates to documentation, including additional references where relevant.

- R
Published by andy-iskauskas about 2 years ago

hmer - v1.5.9

Bug Fixes

• emulator_from_data now safely handles/terminates if users supply width-zero range(s), or if regression is a 'perfect fit' for data; behaviour of check.ranges=TRUE modified to ensure consistent range calculations.
• Calculation of ranges in slice_sample now stable against strange edge cases.
• Minor fixes to some misbehaving unit tests, and for some correlation function behaviour in seldom-used situations. # Enhancements
• seek_good and importance_sample modified to more efficiently propose points where plausible_set is already a large data frame.
• hit_by_wave modified to take n.sig as a parameter, to allow user determination of what constitutes a match to observed data.
• plot_lattice can be used to plot parameter subsets, while maintaining the analysis across all parameters, aiding visualisation for large-dimensional parameter spaces.
• Combined legends can be plotted when passing emulator_plot a small (between 2 and 4) list of emulators using include_legend = TRUE.
• To ensure compatibility with advanced usage of Proto_emulator, generate_new_design can have the emulator ordering overriden.
• Code refactoring and restructuring, including removal of extraneous calls to imported packages for efficiency gains.
• Documentation updates, spellchecks, and fixes.

- R
Published by andy-iskauskas about 2 years ago

hmer - v1.5.6

hmer 1.5.6

Major changes

• New functionality: diagnostic_pass for automated diagnostics and modifications of emulators; hit_by_wave for visualisation of history matching progress.

Bug fixes

• Fixed issue in simulator_plot et al. where some combinations of wave_numbers and zero_in caused out-of-index errors.
• Edge case fixes for training individual variance emulators or one-dimensional variance emulators.

Enhancements

• Vastly improved efficiency of emulator training when emulator_type = "variance" or emulator_type = "covariance".
• Improvement to hyperparameter estimation when training covariance emulators.

- R
Published by andy-iskauskas almost 3 years ago

hmer - v1.5.0

hmer 1.5.0

Major changes

• Change to nomenclature: all xx_emulator_from_data functions are now a single function; variance_emulator_from_data and bimodal_emulator_from_data are now called using emulator_from_data with argument emulator_type = 'variance' and emulator_type = 'multistate' respectively. To avoid confusion about the output of the function, generate_new_runs has been renamed to generate_new_design. Older functions have been deprecated and will be removed in a subsequent version.

Bug fixes

• Changes to generate_new_design to avoid implausibility asymptoting in edge cases.

• Explicit checks included to ensure that ordering of parameter ranges matches with data.frames provided for training

• Fixes to ensure that 1d systems behave as expected.

• Fix to avoid singular correlation matrices in some situations where corr_type = 'exp_sq'.

• Fixes for edge-case multistate emulator training problems, and for situations where model.matrix behaviour fails (due to deparse truncation issues in core R functions).

Enhancements

• Modifications to emulator design via emulator_from_data: in particular hyperparameter estimation has been made more robust; emulator regression surfaces now support cubic terms; variance and multistate emulators now more robust to different numbers of repetitions at different input sites.

• Covariance emulation introduced (via emulator_from_data(..., emulator_type = 'covariance')), allowing a full covariance matrix to be robustly emulated. Results are presented using a prototype emulator matrix EmulatedMatrix to efficiently make predictions, including checks to ensure that predicted matrices are semi-positive definite. In the future, covariance matrices will be supported in implausibility measures and point proposals.

- R
Published by andy-iskauskas about 3 years ago

hmer - v1.4.0

hmer 1.4.0

Bug fixes

• Multiple fixes to deal with singletons: ensuring 1d examples work with generate_new_runs; single output systems behave appropriately under clustering, diagnostics, and implausibility for bimodal_emulator_from_data; modifications to simulator_plot for single output systems; bug-fix in generate_new_runs for user-provided single-element plausible sets.

• Small fixes to guarantee compatibility with new functionality.

• Fixed scoping issue with variance emulators where emulator would not get the correct s_diag function when initialised within a collection of emulators in variance_emulator_from_data

Enhancements

• Beeswarm plots added as alternative to simulator_plot (credit to tjmckinley)

• Custom multi-emulator implausibility now supported within generate_new_runs: multiple conditions can be supplied as part of the point-screening process via accept_measure. Structure of optional arguments for generate_new_runs has been modified (with backwards-compatibility for older code) - see help file for details.

• Added check to emulator_from_data to handle mismatched input names and ranges

New Functionality

• Proto_emulator introduced: the hmer framework of diagnostics, visualisation, and point proposal can be used with entirely custom objects.

- R
Published by andy-iskauskas over 3 years ago

hmer - v1.2.0

hmer 1.2.0

Bug fixes

• Fix to implausibility for variance emulators to take account of ensemble size

• Functions full_wave, variance_emulator_from_data, bimodal_emulator_from_data all accept an na.rm argument to handle missing data

• Fixed edge cases where generate_new_runs could get stuck at a particular implausibility cutoff, and increased stability of termination for points generation from variance/bimodal emulators

• Other small fixes, including modification to ggplot functions to handle deprecation of size aesthetic

Enhancements

• Optimisation of emulator calculations, particularly within correlation matrices

• Optimised point generation to leverage Latin Hypercube Designs, where useful

• Modification to standard_errors to highlight points of interest

• Modifications made to faciliate support for custom emulators and implausibility measures (to come in a later update).

- R
Published by andy-iskauskas over 3 years ago

hmer - v1.0.1

hmer 1.0.1

Bug Fixes

• Implemented more error catching/handling for correlation functions, including explicit stop calls if hyperparameters are not provided/ill-specified and automatic coercion to data.matrix to ensure compatibility with derivative functions

• Internal functions collect_emulators, scale_input, convertRanges and multiply_function modified to handle various edge case usages and issues with multiple waves of stochastic or bimodal emulators

• Modified lhs_gen_cluster so that if emulators cannot/need not be clustered, default lhs_gen behaviour is used

• Fixed Emulator code to address calculation issues with derivatives that meant that partial derivatives did not commute

• Various other small fixes

Enhancements

• Unit testing framework implemented and tracked; unit tests for all relevant functions are now in place

• Functions generate_new_runs and nth_implausible will try to determine a sensible value of n or nth when calculating nth-maximum implausibility if none is supplied by the user.

• Badges and sticker added to readme

Full Changelog: https://github.com/andy-iskauskas/hmer/compare/v.1.0.0...v1.0.1

- R
Published by andy-iskauskas about 4 years ago

hmer - CRAN compliant release

hmer v1.0.0

This is the first version of hmer to be released on CRAN: hmer package.

More improvements and functionality will be coming in the near future!

- R
Published by andy-iskauskas about 4 years ago