.. _method-efficient_global-batch_size: """""""""" batch_size """""""""" Total batch size in parallel EGO .. toctree:: :hidden: :maxdepth: 1 method-efficient_global-batch_size-synchronization method-efficient_global-batch_size-exploration **Specification** - *Alias:* None - *Arguments:* INTEGER - *JSON argument key:* count **Child Keywords:** +-------------------------+------------------------+--------------------+--------------------------------+------------------------------------------------------------+ | Required/Optional | Description of Group | JSON Group Key | Dakota Keyword | Dakota Keyword Description | +=========================+========================+====================+================================+============================================================+ | Optional | | | `synchronization`__ | Select how Dakota schedules a batch of concurrent function | | | | | | evaluations in a parallel algorithm | +-------------------------+------------------------+--------------------+--------------------------------+------------------------------------------------------------+ | Optional | | | `exploration`__ | Portion of batch size dedicated to exploration in parallel | | | | | | EGO | +-------------------------+------------------------+--------------------+--------------------------------+------------------------------------------------------------+ .. __: method-efficient_global-batch_size-synchronization.html __ method-efficient_global-batch_size-exploration.html **Description** Refinement candidates are generated by an acquisition function such as maximum expected improvement, which balances exploration and exploitation. Refinement candidates can also be generated by purely explorative metrics such as maximum prediction variance. For a specified ``batch_size``, ``exploration`` specifies the subset of this total that will be dedicated to pure exploration of the parameter space. *Default Behavior* All of the batch size is devoted to the standard acquisition approach, balancing exploration and exploitation. **Examples** .. tab:: Freeform .. code-block:: dakota method efficient_global seed = 1237 batch_size = 8 exploration = 2 .. tab:: JSON .. code-block:: json { "method": [ { "efficient_global": { "seed": 1237, "batch_size": { "count": 8, "exploration": 2 } } } ] }