Generated from UML->XMI->intermediate->XSD. This class represents the execution of simulation codes. Its protocol is therefore a SimDM/Simulator iso a general SimDM/Protocol. It extends SimDM/Experiment by adding descriptions of the physical processes that were simulated _12_1_8e0028f_1173260098343_801274_269 This class allows users to provide a statistical characterisation of a collection of objects in a specified SimDM/OutputDataset. It represents both 'a priori' and 'a posteriori' characterisations. With a priori characterisation we indicate possible and/or nominal [?] values the variable may take, it defines the possible range of values of the SimDM/Property. In contrast, an a posteriori characterisation of a property in an collection provides summarising, likely statistical, information on the values that were actually measured (i.e. observed, simulated) by the objects in the collection. The a priori characterisation is most similar, in fact a generalisation of the Characterisation model of the IVOA Data Model working group. In the current model we stick to simple (numerical) quantities for characterising a collection of objects. For example the equivalent value of a support from the CharacterisationDM is absent, as it is not terribly useful for discovery and querying, even more so of course for concepts equivalent to sensitivity. _12_1_8e0028f_1173261097203_650221_811 This attribute indicates which statistic is used to statistically summarise the referenced SimDM/ObjectType's SimDM/Property. _12_1_2_213004e4_1207895986906_98049_390 If 'true', this attribute indicates that the statistical characterisation of a SimDM/Propertyis aPriori. That is it indicates that the value says something about what the publisher believes are the possible values the SimDM/Property can assume. An "p priori" characterisation is therefore a translation of the effects that the input configuration (SimDM/ParameterSetting, SimDM/InputDataset), the "provenance-of-the-actual-experiment" therefore) is EXPECTED to have on the final result, represented by the SimDM/OutputDataset-s and the SimDM/Property-s of their constituent objects. Alternatively, if the value is "false", it indicates that the statistical charecterisation is 'a posteriori', i.e. it says something about the actual values that were assumed. This requires in general some level of post processing beyond simply storing the result, but in general using simple statistics such as mean and standard deviation that are not modelled as a separate SimDM/Experiment. _12_1_2_213004e4_1207897770500_675578_405 Summarising value of a numerical SimDM/Property in a collection of objects _12_1_8e0028f_1173261307734_451647_1091 In case a string-typed property is summarised this attribute should be used to give the value. This would likely be applicable to a restricted set of statistics such as nominal value. _12_1_213004e4_1274427320984_956970_413 The SimDM/Property of the SimDM/ObjectType that is being characterised. In the IVOA DM's characterisation model this is represented by specified objects, such as spatialaxis, timeaxis etc. Here we do not know in advance which kind of property is characterised, hence this explicit reference. _12_1_8e0028f_1210838650468_163317_201 This class represents 1 or more instances of a single SimDM/OutputDataObjectType-s produced by an SimDM/Experiment. No assumption is made how a SimDM/OutputDataset is represented in the real world. It may consist of one or multiple files, or possibly one or more tables in a database, or subsets thereof. The exact storage is not currently modelled. The class represents an actual collection of objects without (necessarily) listing these all. Instead it is to be used to allow statistical summaries/characterisation of these products. A special feature is that a SimDM/OutputDataset may define a dependency on another, 'parent' data set. This reflects to possibility in the SimDM/Protocol to model relationships between different SimDM/OutputDataObjectType-s. This is implemented using two references, one to the parent SimDM/OutputDataset, the other to the SimDM/Relationship. _12_1_8e0028f_1173261164953_285409_884 Gives the number of objects in this SimDM/OutputDataset _12_1_8e0028f_1175773837250_855594_253 Represents an optional reference to a URL from which the SimDM/OutputDataset can be obtained. No statement is made on whether this should represent a simple file for download, or a link to a web page with information how to obtain it. _12_1_213004e4_1274417928031_716609_385 This reference to SimDM/DataObjectType indicates the type of data object stored in this collection. _12_1_8e0028f_1330008564000_341998_755 The SimDM/Experiment represents the execution of a computer program represented by a SimDM/Protocol. It defines how the program was run by providing values to parameters for example. It describes the goal of the experiment in terms of the physical objects and processes that are being simulated and it contains the collection of results. _12_1_8e0028f_1175592662265_484162_263 The date/time at which the experiment was completed _12_1_8e0028f_1173262402406_325355_1300 Reference to the Protocol used in the exeuciton of this Experiment. Subclasses of this Experiment will in general override/"subset" this reference to subclasses of Protocol. _12_1_1_8e0028f_1178627594125_954476_600 Represents an experiment that corresponds to the execution of a PostProcessor protocol. It manipulates a pre-existing result to produce new results, but without introducing new physics. _12_1_8e0028f_1175776635250_448091_386 The primary experiment whose results are being post-processed by this PostProcessing step. Simplifies looking up the original experiments, but is in principle redundant with the inputData. _12_1_8e0028f_1315839240812_457040_458 Many experiments require pre-existing data sets for their execution. This class represents such an association for its parent SimDM/Experiment. It is assumed to contain 1 or more objects of a specified SimDM/InputDataObjectType defined on the SimDM/Protocol used by the container SimDM/Experiment. In those cases where the actual input data set is represented by a SimDM/OutputDataset, for example in a SimDB, we can represent the input data by a corresponding reference. Whereas that would be optimal, it may not always be practical. In cases where this is not so, this reference is replaced with an accessURL attribute on this class. This would allow a user to find out whether a product exists in the database with that same url, make the link indirect but arguably more correct. _12_1_8e0028f_1175776701562_299478_435 Describes the role that the input data set plays in the experiment. _12_1_8e0028f_1240401334281_915121_1083 URL by which the input data set can be obtained. _12_1_213004e4_1305783587937_349349_219 The SimDM/OutputDataset produced by an earlier SimDM/Experiment that is used as input data for the current one. _12_1_213004e4_1274347266765_386368_631 Reference to the type definition for this SimDM/InputDataset. This must refer to a SimDM/InputDataObjectType defined on the SimDM/Protocol according to which the SimDM/Experiment is performed. _12_1_213004e4_1305784991671_735859_566 To run an Experiment, one usually needs to assign values to parameters defined on the corresponding Protocol. This class makes the association between the Experiment and the parameters, indicating which parameter is given what value. Input parameters can be given any datatype from the DataType enumeration and hence it is in principle impossible to assign a single datatype to the value attribute storing the parameter setting in this model. Whereas we might use a generic string data type, this will in many implementations limit the use of query expressions that one might use for numerical values for example. As a work around the current design has two attributes, one of type real, one of type string. The former should be used for numerical InputParanmeters, the latter for all others. _12_1_2_213004e4_1195984162343_63311_299 Attribute that holds on to the actual parameter value in case the referenced input parameter is not a numerical type. _12_1_8e0028f_1255612276171_485250_550 Attribute that holds on to the actual parameter value in case the referenced input parameter is a numerical type. _12_1_8e0028f_1255612329015_690320_556 Reference to the actual input parameter whose value is being set. _12_1_8e0028f_1213803596859_993272_300 The AppliedPhysics class represents the association between a physical process on a Simulator and this class' parent Simulation. Many simulation codes allow one to turn on or off modules corresponding to different physics. For example certain SPH codes allow one to turn off the hydrodynamics, leaving only gravity as the physical process being simulated. _14_0_8e0028f_1202819493156_452128_1609 Reference to the Simulator's Physcis module that is used in the Simulation. _14_0_8e0028f_1202819514125_954655_1685 The AppliedAlgorithm class represents the application of a particular algorithm in an experiment. Some simulation codes allow one to choose between different algorithms for representing a particular process. For example some N-Body codes allow one to choose between Tree-only or Tree+ParticleMesh codes. To indicate which algorithms were actually used in an experiment one adds an instance of this AppliedAlgorithm class to the collection on the experiment. _12_1_8e0028f_1224862111906_82091_852 Reference to the actual algorithm that is applied _12_1_8e0028f_1224862123421_661003_925 Represents an individual object of a given object type produced by an experiment. Is required if we want to represent object-object collection hierarchies. NB the name DataObject is used iso Object, as the latter may lead to name clashes in serialisations. In this model any artefact produced by a Protocol can be represented by an appropriate ObjectType and relations can exist between them. Some of these relations are many-to-one, i.e. between object collections and parent ojects. If such relations must also be represented in the output of an Experiment, it must be possible to represent individual objects. A typical example is a cosmological simulation produccing N > 1 snapshots. We may want to represent each individual snapshot in the result, together with statistical information about the collections of particles contained in each. The object can give values to properties related to objects of its type. In principlpe this model now would allow one to describe an experiment in all details, with each individual object listed. This is as should be for a conceptual model, but in particular applications we envision that limits be put on the size of results. For example it does not make sense to store the complete results of a large cosmolgical simulation in this way, as the storage of property values is inefficient. _12_1_213004e4_1323152778562_515768_231 Associates an individual Object to an InputDataSet. Thus if for example only 3 individual snapshots out of a collection of N snapshots produced by a cosmological simulation are post-processed, the actual snapshots used can be indicated using this associative object. _12_1_213004e4_1323154308015_700073_594 Reference to an actual object that is being used in an experiment. _12_1_213004e4_1323154502125_877066_670 Assign a value to a property of a data object. _12_1_213004e4_1323153854062_619645_318 If the Property to which this value is assigned is not numeric, this attribute should be used to represent the value. _12_1_213004e4_1323154039359_335644_388 It the property to which this value is assigned is numerical, use this attribute to represent the value. _12_1_213004e4_1323154084546_613266_393 The property to which the value is assigned. _12_1_213004e4_1323154166765_707449_406 Instance of a composition Relationship between an ObjectType and a collection of objects. _12_1_8e0028f_1323966683328_800469_231 TODO : Missing description : please, update your UML model asap. _12_1_8e0028f_1323966704468_292205_304 reference to the relationship definition on the ObjectType that is represented by the ObjectCollection. The referenced Relationship MUST have relationshipType='composition'. _12_1_8e0028f_1330008511015_22980_742 Instance of a reference Relationship between two ObjectType-s. _12_1_8e0028f_1330002173406_842171_233 The object referenced by the ObjectReference's parent DataObject. _12_1_8e0028f_1330002243609_125184_374 Reference to the relationship definition on the ObjectType that is represented by the ObjectReference. The referenced Relationship MUST have relationshipType='reference'. _12_1_8e0028f_1330008515671_377198_748 This enumeration lists possible statistics that can be used to characterise a collection of Property-s in a result. The literals in this enumeration in general correspond to the result of simple statistical operation on this collection (when a posteriori) or on the corresponding operation on the a priori probability distribtion (assuming this can be defined) of its values. There is overlap between this enumeration and the 'stat' family of UCDs in the UCD controlled vocabulary (http://www.ivoa.net/Documents/REC/UCD/UCDlist-20070402.html). Where this enumeration is used one might consider using a vocabulary based on those terms. _12_1_2_213004e4_1207894335531_849311_215 In case an output data set contains exactly 1 object, the value statistic should be used to indicate the actual property value. A typical value, as judged by the publisher of the results. May not correspond to a predefined statistical operation such as mean or mode. Represents the minimum ('possible' if a priori, 'attained' if a osteriori) value for the characterised property. Represents the maximum ('possible' if a priori, 'attained' if a osteriori) value for the characterised property. Represents the mode of the distribution ('expected' if a priori, 'attained' if a posteriori) for the characterised property. Represents the mean ('expected' if a priori, 'attained' if a posteriori) value for the characterised property. Represents the median of a distribution. Represents the expected/attained standard deviation for the characterised property. Represents the expected/attained variance for the characterised property.