10 Software Framework
10.1 Definitions
Following are definitions of terms used in this chapter:
Model System
A definition for a set of related models and a software framework for implementing that definition. Models built in the modeling system are related by the domains being modeled (e.g. travel, energy consumption, hydrology, etc.), the ‘agents’ being modeled (e.g. households, cities, watersheds, etc.), how physical space is represented (e.g. zones, grids, cubes, etc.), how time is represented (e.g. continuouse vs. discrete, independent vs. dependent on past states), and other modeling goals and tradeoffs (e.g. representational detail, degree of coupling, run times, etc.). The model system definition includes specifications for model modules that can be used in the model system, file structure specifications for organizing model parameters and input data necessary for running a model. The software framework for the model system is a library of code that manages the execution of model modules that are designed to work in the model system.Model
A model as used in this document refers to a model such as GreenSTEP that calculates a number of different attributes (e.g. household size, household income, number of autos owned, vehicle-miles traveled, etc.) that are composed of a number of components (submodels) that each calculate one or a few attributes.Submodel
A submodel is the component of a model that calculates one or a few closely related attributes.Module
A module, at its heart, is a collection of data and functions that meet the specifications described in this document and that implement a submodel. Modules also include documentation of the submodel. Modules as made available to users in R packages. Typically a set of related modules is included in a package.Software Framework
A software framework is a library of code containing functions that manage the execution of modules. These functions manage all interactions between modules, model system variables, and a datastore.Datastore
A datastore is a file or set of files for storing all of the inputs used by modules and outputs produced by modules.
10.2 Model System Objectives
The GreenSTEP model and related models are disaggregate strategic planning models. They are disaggregate because, like many modern transportation models, they simulate behavior at the individual household level rather than at a more aggregate ‘zonal’ level. This enables the assessment of how prospective policies or other changes could have different impacts on different types of households (e.g. low income vs. high income). The models are strategic planning models because they are built to support long-range strategic planning decisions such as community visioning, policy development, and scenario planning. Strategic planning processes most often need to consider a number of possibilities about how the future may unfold and a range of potential actions that might be taken. As a consequence, models built to support strategic planning need to be responsive to a large number of variables and be capable of running quickly so that a large number of runs can be done to explore the decision space. The VisionEval model system supports the development of these types of models. The design objectives for this model system are:
Modularity
The model system will allow new capabilities to be added in a plug-and-play fashion so models can be improved and extended and so improvements developed for one model can be easily shared with other models. Models are composed of modules that contain all of the data and functionality needed to calculate what they are intended to calculate.Loose Coupling This objective is closely related to the modularity objective. Loose coupling is necessary if modules are to be added to or removed from models in a plug-and-play fashion. Loose coupling means that the parameter estimation for a submodel is independent of the parameter estimation of any other submodel. It also means that dependencies be modules are well defined and minimized. Most communication between modules is carried out through the transfer of data that is mediated by the software framework.
Openness
The VisionEval software framework and all modules developed to operate in the framework will be completely open. Being open means more than sharing ones work. It means completely revealing ones work so that others can assess how the module works. All module code, parameters, data, and specifications will be open to inspection and licensed using an open source license (e.g. Apache 2) that allows users to use, modify, and redistribute as they see fit. In addition, modules will provide access to data and code to estimate the model that the module implements. Finally, a module will contain complete documentation that users may use to document the model that the module is a part of.Geographic Scalability
The model system will enable models to be applied at a variety of geographic scales including metropolitan areas of various sizes, states of various sizes, and multi-state regions. Although models are applied at different scales, they share common geographic definitions to enable modules to be more readily shared between models built for the modeling system.Data Accessibility
Model results will be saved in a datastore that is easy to query. Results can be filtered, aggregated, and post-processed to produce desired performance measures.Regional Calibration Capability
Modules will have built in capbilities for estimating and calibrating submodel parameters from regional data when necessary.Speed and Simplicity
Since the intent of the model system is to support the development of strategic planning models, it is important that the models be able to address a large number factors and be able to model a large number of scenarios. For this to occur, the framework needs to run efficiently and modules need to be simple and need to run quickly.Operating System Independence The model system will run on any of the 3 major operating systems; Windows, Apple, or Linux. As is the case with GreenSTEP and related models, the VisionEval model system is written in the R programming language. Well-supported and easily installed R implementations exist for these operating systems. Modules will be distributed as standard R packages that can be compiled on all operating systems. Code that is written in another language may be included in a module package as long as it can be compiled in an R package that is usable on all 3 operating systems.
To help ease maintenance of VE, it is recommended to minimize the use of new R libraries, and to reuse R libraries used by existing VE modules when possible.Preemptive Error Checking
The model system will incorporate extensive data checking to identify errors in the model setup and inputs at the very beginning of the model run. Error messages will clearing identify the causes of the errors. The objective of early error checking is to avoid model runtime errors that waste model execution time and are difficult to debug.
10.2.1 Model System Software Design Approach
The VisionEval software framework uses many of the ideas from the functional programming paradigm to create a modeling system that is modular and robust. The framework is implemented in the R programming language which is largely a functional programming language having a well developed system for managing packages of modules. This section describes how the VisionEval model system incorporates functional programming design ideas and the reason for using the R programming language for implementing those ideas.
10.2.1.1 Functional Programming Inspirations for System Design
There are a number of programming paradigms for organizing software and implementing modular system designs. The VisionEval system design is most inspired by the functional programming paradigm. This section describes key characteristics of the functional design paradigm that influence the VisionEval model system design to achieve the design objectives.
The functional programming paradigm views computing as sequences of data transformations where a functions successively process a data stream with the outputs of one function becoming the inputs of the next. Functions are data processing machines with defined inputs and outputs. Given the same set of inputs, functions always produce the same set of outputs. Although functions have requirements for the data they process, they are not bound to the data in the same way as functions (methods) in the object-oriented programming paradigm. This characteristic of the functional paradigm fits well with the ‘loose coupling’ objective for the model system where model modules only interact with one another through the transfer of data mediated by the framework software. The design is for each module to act like a function with the framework software calling it, providing the data the module needs, and saving the data that the module produces.
Another key characteristic of the functional programming paradigm is that functions should have no side effects. A function should only change the state of variables within it’s scope and should have no effects on the system outside of its scope. The only way that a function interacts with the rest of the program is by returning the result of its calculations to whatever function called it. Such functions are called pure functions. This characteristic makes software more testable, reliable, and maintainable. It is key aspect of the VisionEval model system design. The system is designed so that modules act like pure functions. The only thing that they do is return the results of their calculations to the software framework. They do not make any changes to program variables outside their scope and they do not read from or write to files. This approach makes the framework very robust and modular because code within a module can be changed without affecting any other module.
The framework software itself is also designed to minimize side effects in order to make it more robust and maintainable. Global state variables for a model run are kept to a minimum. Almost all model run state information is kept in the ‘ModelState.Rda’ file (Section 6.6). A common datastore holds all of the results of model computations. The only side effects in the framework code are reading model definition and input files, reading and writing to the common datastore, and writing to a log file.
A third key characteristic of some functional programming languages is the extensive use of data typing. While compiled languages in general use data typing to check the properness of functions and expressions when a program is compiled, some functional programming languages like Haskell and Elm make extensive use of defining and checking different data types for function inputs and outputs. They also include a type notation system for documenting functions. This makes it easier to check and understand the code. An analogous approach is used in the VisionEval system design. Each module includes specifications for all data that it consumes and all data that it produces. This enables the framework software to check that modules will work properly together and enables a model to be checked thoroughly before it is run so to eliminate run time errors. It also clearly documents to others what data the module uses and what it produces.
10.2.1.2 Use of the R Software Environment to Implement the Model System
The VisionEval model system is built in the R programming language for statistical computing and graphics. R is an open-source variant of the S programming language developed at Bell Labs but is more functional in nature than S. Although R was primarily developed to be an interactive programming environment for data analysis, the language has a full set of features that enables it to be used for all steps in the modeling process from data preparation and cleaning through model implementation and output visualization. The language is augmented by thousands of packages supporting data analysis, programming, and visualization. The interactive nature of the language, range of capabilities, and large number of supporting packages enabled the the GreenSTEP model to be developed in an agile manner in a relatively short period of time. At the time, no other programming language had this range of capabilities and the large number of supporting packages. The VisionEval model system uses the R language for the following reasons:
1) The existing code base for the GreenSTEP model and related models is written in R. Writing the VisionEval software framework in R enables this code base to be moved to the new framework with much less effort than would be required if it had to be rewritten in another programming language.
2) R is open-source software that is available on all major operating systems so the model system will be operating system independent.
3) R has a very good and well tested package system for packaging modules that is well supported with documentation and build tools. The package system and development tools also include easy-to-use capabilities for documentation, including literate programming. This simplifies the development of the software framework and simplifies the process for module developers to produce complete and well documented modules.
4) R has the most extensive set of statistical and other data analysis packages available. Because of this, almost any type of model can be estimated using R and therefore, modules can contain not only full documentation of model estimation, but also scripts that allow model estimation to be replicated and rerun using regional data.
5) R is an interpreted language with capable (and free) integrated development environments. Because the state of objects can be easily queried, the process of building and testing models is simplified. This makes it easier for modelers who don’t come from a computer science background to develop models to be deployed in the model system.
6) Although as an interpreted language, R is slower than compiled languages, most of the core functions are “vectorized” functions that are written in C. This means that R programs can carry out many operations very quickly. In addition, it is relatively easy to call functions written in compiled languages such as C++, C, and Fortran to R so that if a pure R model is not fast enough, portions can be written as functions in a compiled language and linked to the R code.
7) R has a large user base and so it is relatively easy for users to get answers to programming questions.
10.3 Model System Layers
The VisionEval model system is composed of 3 layers:
1) Model: The model layer defines the structure of the model and organizes all of the modules into a coherent model. The model layer includes a module run script, model definition files, model input files, and common datastore.
2) Modules: The module layer is the core of a model. Modules contain all of the code and parameters to implement submodels which are the building blocks of models.
3) Software Framework: The software framework layer provides the functionality for controlling a model run, running modules, and interacting with the common datastore.
These layers are illustrated in Figure 1. Following sections describe the design and specifications for each layer.
Figure 1. Overview of VisionEval Model System
A VisionEval model is built from a set of compatible modules, a set of specifications for the model and geography, a set of scenario input files, and a simple R script that initializes and runs the model. Following is a simple example of a model script:
#Initialize and check the model
initializeModel(
ParamDir = "defs",
RunParamFile = "run_parameters.json",
GeoFile = "geo.csv",
ModelParamFile = "model_parameters.json",
LoadDatastore = FALSE,
DatastoreName = NULL,
SaveDatastore = TRUE
)
#Run modules for all forecast years
for(Year in getYears()) {
runModule(
ModuleName = "CreateHouseholds",
PackageName = "SimHouseholds",
RunFor = "AllYears",
RunYear = Year)
runModule(
ModuleName = "PredictWorkers",
PackageName = "SimHouseholds",
RunFor = "AllYears",
RunYear = Year)
runModule(
ModuleName = "PredictLifeCycle",
PackageName = "SimHouseholds",
RunFor = "AllYears",
RunYear = Year)
runModule(
ModuleName = "PredictIncome",
PackageName = "SimHouseholds",
RunFor = "AllYears",
RunYear = Year)
...
}
A full model run script is shown in Appendix A.
The script calls two functions that are defined by the software framework; initializeModel and runModule. The initializeModel function initializes the model environment and model datastore, checks that all necessary modules are installed, and checks whether all module data dependencies can be satisfied. The arguments of the initializeModel function identify where key model definition data are found. The initializeModel function allows users to load an existing datastore. This functionality is useful for scenario management, retaining and building upon the results of previous model runs. For example, a set scenarios could be managed in the following manner using the capability for loading datastores:
- A base year model is calibrated and validated
- Future land use scenarios are developed and run (e.g. 3 different scenarios having different growth and development assumptions). Each land use scenario loads the datastore for the base year model.
- Future transportation policy scenarios are developed and run for each land use scenario (e.g. 3 different scenarios having different policy assumptions). Each transportion policy scenario loads the datastore for a land use scenario model run.
The runModule function, as the name suggests, runs a module. The arguments of the runModule function identify the name of the module to be run, the package the module is in, and whether the module should be run for all years, only the base year, or for all years except for the base year. This approach makes it easy for users to combine modules in a ‘plug-and-play’ fashion. One simply identifies the modules that will be run and the sequence that they will be run in. This is possible in large part for the following reasons:
1) The modules are loosely coupled. Modules only communicate to one another by passing information to and from the datastore or by calling the services of another module. Module calling is described in detail in section 8.1.2.2.
2) The framework establishes standards for key shared aspects of modules including how data attributes are specified and how geography is represented.
3) Every module includes detailed specifications for the data that are inputs to the module and for the data that are outputs from the module. These data specifications serve as contracts that the framework software enforces.
The runModule function and how it is implemented by the software framework allows developers and advanced users to create different module variants and save them in a package having a different name. Users can then identify the alternative package in the runModule function call to use the variant version. For example, the VEPowertrainsAndFuels package includes data and modules that model key vehicle powertrain characteristics such as powertrain type proportions (ICEV, HEV, PHEV, BEV) by vehicle type and vehicle model year. A developer or advanced used could create package version that reflects Energy Information Agency (EIA) projections and name it VEPowertrainsAndFuelsxEIA, and another version which reflects California zero emissions vehicle standards and name it VEPowertrainsAndFuelsxCAZEV. The user could then specify which package version they want to run in the runModule function call. This functionality also allows developers and advanced users to test out module changes without altering an ‘official’ package.
These features and how they are designed are described in detail in following sections.
10.4 Model Layer Description
The model layer is composed of:
- The directory (i.e. folder) and file structure for organizing scenario inputs and model parameters;
- Model parameter files describing the model geography (consistent with standard definitions) and global parameters;
- The model run script that lists the model execution steps; and,
- The datastore which stores all of the data produced during the execution of the model.
Each of these components is described in the following subsections.
10.4.1 Model Directory Structure
A model application has a very simple directory structure as shown in the following representation of a directory tree.
my_model
| run_model.R
| <ModelState.Rda>
| <logXXXX.txt>
| <datastore>
|
|
|____defs
| | run_parameters.json
| | model_parameters.json
| | geo.csv
| | units.csv
| | deflators.csv
|
|
|____inputs
| filename.csv
| filename.csv
| ...
The overall project directory, named my_model in this example, may have any name that is allowed by the operating system the model is being run on. One file is placed in this top level directory by the user, “run_model.R.” Three additional files, denoted in the diagram by angled brackets, are created in the course of checking and running the model.
The “run_model.R” file, introduced in the previous section, initializes the model environment and datastore, checks that all necessary packages are installed, checks whether data dependencies can be satisfied, and then runs modules in a specified sequence. Data checks are performed before any modules are run to catch any errors. This saves time and the aggravation that occurs when a model run fails in midstream due to incorrect data inputs or other errors due to incorrect model setup. Data checking in advance is possible because every module includes detailed specifications for its input and output data. All scenario input files are checked against specifications to determine whether the required data exist and are correct. In addition, the state of the datastore is ‘simulated’ in the order that each module will be run to determine whether the data each module needs will be available in the datastore. After the model has been initialized and all all data checks are satisfactory, the modules are executed in the sequence prescribed in the script.
The “ModelState.Rda” file is a R binary file that contains a list that holds key variables used in managing the model run. The file is created when the model run is initialized and is updated whenever the state of the datastore changes. Framework functions read this file when necessary to validate data and to determine that datastore read and write operations can be completed successfully. This file is described in more detail in section 6.6.
The “logXXXX.txt” file is a text file that is created when the model is initialized. This log file is used to record model run progress and any error or warning messages. The ‘XXXX’ part of the name is the date and time when the log file is created.
The “datastore” is a file or directory that contains the central datastore for the model. The VisionEval framework supports multiple types of datastore. Currently, two types are supported. The preferred type (designated RD) stores datasets in the form of native R data files. Files are stored in a hierarchical directory structure with the top-level directory named “Datastore.” Another type (designated H5) stores all data in a single binary HDF5 file. The logical structure of these two datastore types are very similar and are described in detail below. Users specify the datastore name in the “parameters.json” file (see below).
The “defs” directory contains all of the definition files needed to support the model run. Five files are required to be present in this directory: “run_parameters.json,” “model_parameters.json,” “geography.csv,” “deflators.csv,” and “units.csv.”
The “run_parameters.json” file contains parameters that define key attributes of the model run and relationships to other model runs. The file is a JSON-formatted text file. The JSON format is used for several reasons. First, it provides much flexibility in how parameters may be structured. For example a parameter could be a single value or an array of values. Second, the JSON format is well documented and is very easy to learn how to use. It uses standard punctuation for formatting and, unlike XML, doesn’t require learning a markup language. Third, files are ordinary text files that can be easily read and edited by a number of different text editors available on all major operating systems. There are also a number of commercial and open source tools that simplify the process of editing and checking JSON-formatted files.
The “run_parameters.json” file specifies the following parameters:
Model The name of the model. Example: “Oregon-GreenSTEP.”
Scenario The name of the scenario. Example: “High-Gas-Price.”
Description A short description of the scenario. Example: “Assume tripling of gas prices.”
Region The name of the region being modeled. Example: “Oregon.”
BaseYear The base year for the model. Example: “2015.”
Years An array of all the ‘forecast’ years that the model will be run for. Example: [“2025,” “2050”].
DatastoreName The name of the datastore. It can be any name that is valid for the operating system. It is recommended that this be named “datastore.h5” for HDF5 datastores and “Datastore” for R data file datastores.
DatastoreType A 2-letter abbreviation identifying the datastore type: “RD” for R data file datastore, “H5” for HDF5 datastore. The framework uses the DatastoreType abbreviation to choose the functions used to initialize the datastore and interact with it.
Seed This is a number that modules use as a random seed to make model runs reproducible.
RequiredVEPackages Lists all of the VisionEval packages that contain modules that are called by the model. Note that the required packages that are listed must be consistent with the packages that are identified in the run_model.R script.
The “model_parameters.json” can contain global parameters for a particular model configuration that may be used by multiple modules. For example, a model configuration to be a GreenSTEP model may require some parameters that are not required by a model configuration for an RSPM model. Parameters in this file should not include parameters that are specific to a module or data that would more properly be model inputs. While this file is available to establish global model parameters such as the value of time, it should be used sparingly in order enhance transferrability of modules between different models.
The “geography.csv” file describes all of the geographic relationships for the model and the names of geographic entities in a CSV-formatted text file. The CSV format, like the JSON format is a plain text file. It is used rather than the JSON format because the geographic relationships are best described in table form and the CSV format is made for tabular data. In addition, a number of different open source and commercial spreadsheet and GIS programs can export tabular data in a CSV-formatted files. The structure of the model system geography is described in detail in Section 6.2 below.
The “units.csv” file describes the default units to be used for storing complex data types in the model. The VisionEval model system keeps track of the types and units of measure of all data that are processed. The model system recognizes 4 primitive data types, a number of complex data types (e.g. currency, distance), and a compound data type. The primitive data types are data types recognized by the R language: ‘double,’ ‘integer,’ ‘character,’ and ‘logical.’ The complex data types such as ‘distance’ and ‘time’ define types of data that have defined measurement units and factors for converting between units. The compound data type combines two or more complex data types whose units are defined in an expression (e.g. MI/HR where MI is the complex unit for miles and HR is the complex unit for hours). The units.csv describes the default units used to store complex data types in the datastore. The file structure and an example are described in more detail in Section 6.3 below.
The “deflators.csv” file defines the annual deflator values, such as the consumer price index, that are used to convert currency values between different years for currency demonination. The file structure and an example are described in more detail in Section 6.4 below.
The “inputs” directory contains all of the input files for a scenario. All input files are CSV-formatted text files. Each module specifies what input files it needs and names and types of data to be included in the needed files. There are several requirements for the structure of input files. These requirements are described in section 6.5 below.
10.4.1.1 Model Geography
The design of the model system includes the specification of a flexible standard for model geography in order to fulfill the objectives of modularity and geographic scalability. As a standard, it specifies levels of geographical units, their names, their relative sizes, and the hierarchical relationships between them. It is flexible in that it allows geographical boundaries to be determined by the user and it allows the units in some geographical levels to be simulated rather than being tied to actual physical locations. Allowing simulation of one or more geographic levels enables modules to be shared between models that operate at different scales. For example a statewide model and a metropolitan area model could use the same module for assigning households to land development types even though the statewide model lacks the fine scale of geography of the metropolitan model.
Following is the definition of the geographic structure of the VisionEval model system:
Region
The region is the entire model area. Large-scale characteristics that don’t vary across the region are specified at the region level. Examples include fuel prices and the carbon intensities of fuels.Azones
Azones are large subdivisions of the region containing populations that are similar in size to those of counties or Census Public Use Microdata Areas (PUMA). The counties used in the GreenSTEP and EERPAT models and metropolitan divisions used in the RSPM are examples of Azones. Azones are used to represent population and economic characteristics that vary across the region such as demographic forecasts of persons by age group and average per capita income. The Azone level of geography must represent actual geographic areas and may not be simulated.-
Bzones
Bzones are subdivisions of Azones that are similar in size to Census Block Groups. The districts used in RSPM models are examples of Bzones. Bzones are used to represent neighborhood characteristics and policies that may be applied differently by neighborhood, for example in the RSPM:- District population density is a variable used in several submodels;
- An inventory of housing units by type by district is a land use input; and,
- Carsharing inputs are specified by district.
In rural areas, Bzones can be used to distinguish small cities from unincorporated areas.
Bzones may correspond to actual geographic areas or may be simulated. Bzone simulation greatly reduces model input data requirements while still enabling the modeling of land-use-related policies and the effects of land use on various aspects of travel behavior. In VE-RPAT models, Bzones are simulated as place types which characterize the intensity and nature of development. In VE-State models, Bzones are synthesized to represent characteristics likely to be found in an actual set of Bzones within each Azone.
- District population density is a variable used in several submodels;
Mareas
Mareas are collections of Azones associated with an urbanized area either because a portion of the urbanized area is located in the Azone or because a substantial proportion of the workers residing in the Azone work at jobs located in the urbanized area. Metropolitan models typically only have one assigned Marea whereas state models may have several. The model system requires that each Azone may be associated with only one Marea. It is also required that all Azones be associated with an Marea. A special Marea named ‘None’ is used to apply to Azones that are not associated with any urbanized area. Mareas are used to specify and model urbanized area transportation characteristics such as overall transportation supply (transit, highways) and congestion. They are also used to specify large scale land-use-related characteristics and policies in models that use Bzone synthesis.
Geographical relationships for a model are described in the “geography.csv” file contained in the “defs” directory. This file tabulates the names of each geographic unit (except for Region) and the relationships between them. Each row shows a unique relationship. Where a unit of geography is not explictly defined (i.e. it will be simulated), “NA” values are placed in the table. Appendix B shows examples of the “geography.csv” file where only Azones are specified and where Azones and Bzones are specified. It should be noted that there are no naming conventions for individual zones. The user is free to choose what conventions they will use.
10.4.1.2 6.3. Data Types, Units, and Currency Deflators
A key feature of the VisionEval model system that enables modules to be bound together into models is a data specifications system. All datasets that a module requires as inputs and datasets that that a module produces must be specified according to requirements. Section 8 describes these specifications in more detail. This section provides an introduction to the TYPE and UNITS specification requirements to provide context for understanding the “units.csv” file in the “defs” directory.
The TYPE attribute for a dataset identifies the data type. The UNITS specification identifies the units of measure. The TYPE and UNITS specifications are related. The TYPE attribute affects the values that may be specified for UNITS attribute and how the framework processes the units values. The model system recognizes 3 categories of data types: ‘primitive,’ ‘complex,’ ‘compound.’ The ‘primitive’ category includes the 4 data types recognized by the R language: double, integer, character, and logical. A dataset that is specified as one of these types has no limitations on how the units of measure are specified. The ‘complex’ category currently includes 13 data types such as currency, distance, and area as shown in the table below. A dataset that is one of these types is limited to specified unit values. For example, the allowed units for the ‘distance’ type are MI, FT, KM, and M (for mile, foot, kilometer, and meter). The ‘compound’ category is composed of the compound data type. For compound data, units are represented as an expression involving the units of complex data types. For example, a dataset of vehicle speeds can be specified as having a TYPE that is ‘compound’ and UNITS that are ‘MI/HR.’ The type is compound because it is made up of two complex types; distance and time. The units are an expression containing distance and time units and the ‘/’ operator. The ’*’ (multiplication) operator may also be used in the units expression for a compound data type. Appendix C documents all the types and units in more detail.
Although the complex and compound data types limit what values the units attributes may have, specifying these types enables the framework software to take care of unit conversions between modules and the datastore. For example, say that a model datastore contains a dataset for Bzone population density having units of persons per square mile. In this case the TYPE attribute for the data would be ‘compound’ and the UNITS would be ‘PRSN/SQMI.’ If a module which calculates household vehicle travel needs population density measured in persons per acre, the module would specify the UNITS as ‘PRSN/ACRE’ and the framework would take care of converting the density values from the units used in the datastore to the units requested by the module. This simplifies matters for module developers and reduces the likelihood of errors due to data conversions.
Although the units specified by a module for a complex data type may be any of the recognized units (e.g. for distance - MI, FT, KM and M), this flexibility does not apply to the datastore. Complex data is stored in the datastore in predefined ways to limit potential confusion and simplify unit conversions. The default units file (units.csv) in the “defs” directory declares the default units to use for storing complex data types in the datastore. This file has two fields named ‘Type’ and ‘Units.’ A row is required for each complex data type recognized by the VisionEval system. The listing to date of complex types and the default units in the demonstration models are as follows:
| Type | Units | Description |
|---|---|---|
| currency | USD | U.S. dollars |
| distance | MI | miles |
| area | SQMI | square miles |
| mass | LB | pounds |
| volume | GAL | gallons |
| time | DAY | days |
| energy | MJ | megajoules |
| people | PRSN | persons |
| vehicles | VEH | vehicles |
| trips | TRIP | trips |
| households | HH | households |
| employment | JOB | jobs |
| activity | HHJOB | households & jobs |
For ‘currency’ data it’s not sufficient to convert values to different units, it’s also necessary to convert currency values between years to account for the effects of inflation. Because the model parameters estimated in different modules may come from datasets collected in different years, and because model users will most likely want to report currency values in current year terms, it is necessary to convert currency values between years. Currency denominated datasets are stored in the datastore as base year values. When a module needs to use a currency denominated dataset, the framework converts base year values to the year values that the module needs. If a module calculates a currency denominated dataset that is to be saved in the datastore, the framework converts those values from the currency year that the module uses to base year values to save in the datastore. The software framework takes care of the process of converting currency values between years automatically and in a consistent manner. This eliminates the need for model developers to convert currency values. It also allows more flexibility for model users and module developers, and for the evolution of the VisionEval model system with new and improved modules, because it eliminates the need to establish a reference year that is to be used for all modules and models.
Modules specify the year of a currency dataset using a modifier to the the UNITS specification. This is done by adding a period and 4-digit year to the specification. For example, the UNITS specification for year 2000 dollars would be ‘USD.2000.’ Note, however, that this convention does not apply to specifications for currency data that is read in from an input file because the model user is free to establish any currency year they choose for currency inputs. In those cases, the year is specified in the input file. This is explained in Section 6.4.
Currency values are converted between years using a deflator series is defined for the model in the “deflators.csv” file in the “defs” directory. This file has 2 columns, ‘Year’ and ‘Value.’ Values are needed for all years specified by modules used in a model in addition to the base year and any other years that currency values in input files will be denominated in. For example, if the modules to be used in a model use dollar denominated values for the years 2000 and 2009, the model base year is 2010, and some input data are denominated in 2015 dollars, then at a minimum the “deflators.csv” file must include deflators for those years. A more flexible approach would be to have an annual series of deflators running from the earliest year through the latest year. Note: it is not necessary to specify deflators for any future model years (e.g. 2030, 2050). All modules in the model system make calculations in constant (uninflated) dollar terms. Deflators are only used to convert user input values to a constant base and to convert values to the year that is consistent with a module’s estimation data.
The UNITS value may also specify a multiplier option for complex and compound data types. This capability exists because modules may use data that are represented in thousands or millions when numbers are very large. For example, freeway and arterial construction costs may be represented in thousands of dollars per mile. A multiplier option is added to a units name by adding a period and then the multiplier expressed in scientific notion where the leading digit must be a 1 (e.g. 1e3 for thousands). For currency units, the multiplier option must follow the year notation. For example, 2010 dollars expressed in thousands would be expressed as ‘USD.2010.1e3.’ Miles traveled expressed in millions would be ‘MI.1e6.’
10.4.1.3 6.4. Model Inputs
The inputs directory contains all of the model inputs for a scenario. A model input file is a table that relates one or more input fields to geographic units and years. Because of the tabular nature of the data, all input files are CSV-formatted text files. The first row of the file contains the headers identifying the data in each column. The columns include each of the data items specified in the input specifications for the module the input file is used for. In addition, the file may be required to have columns labeled Geo and Year depending on which of the following 4 types the input file is:
Inputs apply to the entire region and to all years: In this case, the input file consists of one data row and each column corresponds to a data item.
Inputs apply to parts of the region and to all model years: In this case, the input file consists of one data row for each geographic area and the file must include a column labeled Geo that is used for identifying the geographic areas. For example, if the input file applies to Azones and the model has 10 Azones, the file must have 10 rows in addition to the header. The Geo column identifies each of the Azones. Note that only the geographic areas specified in the geo.csv may be included in this file. If unlisted geographic areas are included, the model run will stop during initialization and the log will contain messages identifying the error(s).
Inputs apply to the entire region but vary by model year: In this case, the input file consists of one data row for each model year and the file much include a column labeled Year that is used for identifying the model years. For example, if the model run parameters specify that the model is to be run for the years 2010 and 2040, the input file must contain 2 rows in addition to the header. The Year column identifies each of the model run years. Note that only the specified model run years may be included in this file. Note that the file may include data for other years in addition to the data for the module run years. When the inputs are processed, the non-model-run year data will be ignored.
Inputs apply to parts of the region and vary by model year: In this case the input file consists of one data row for each combination of geographic area and model year. The file must include a Geo column and a Year column. There must be as many rows as there are combinations of geography and years. For example if an input file applies to Azones and the model specifies 10 Azones and 2 model run years, the file must have 20 rows to accommodate all the combinations in addition to a header row. As with the case described above, inputs may be included for non-model-run years as well as model-run years.
By convention, input file names which include inputs that vary by level of geography, include the level of geography in the input file name. File names should be descriptive. Following are some examples:
- azone_hh_pop_by_age.csv
- azone_hhsize_targets.csv
- bzone_dwelling_units.csv
The name of an input file and the names of all the columns except for the “Geo” and “Year” columns are specified by the module that requires the input data. In addition to specifying the file and column names, the module specifies:
- The level of geography the inputs are specified for (e.g. Region, Azone, Bzone, Czone, Marea);
- The data types in each column (e.g. integer, double, currency, compound);
- The units of the data in each column (e.g. MI, USD); and,
- Acceptable values for the data in each column.
The module section describes these specifications in more detail below. Appendix D shows examples of the two types of input files.
The field names of the input file (other than the “Geo” and “Year” fields) can encode year and unit multiplier information in addition to the name of the data item. This is done by breaking the name into elements with periods (.) separating the elements as follows:
For ‘currency’ data type: Name.Year.Multiplier. For example, TotalIncome.2010.1e3 would be the field name for total income in thousands of 2010 dollars.
For all other data types: Name.Multiplier. For example, TotalDvmt.1e6 would be the field name for total daily vehicle miles traveled in millions.
Where:
Name is the dataset name. This must be the same as specified in the module that calls for the input data.
Year is the four-digit representation of the year that the currency values are denominated for. For example if a currency dataset is in 2010 dollars, the ‘Year’ value would be ‘2010.’ The field name for a currency field must include a ‘Year’ element.
Multiplier is an optional element which identifies the units multiplier. It must be expressed in scientific notation (e.g. 1e3) where the leading digit must be 1. This capability exists to make it easier for users to provide data inputs that may be more conveniently represented with a smaller number of digits and an exponent. For example, annual VMT data for a metropolitan area or state is often represented in thousands or millions.
The the VisionEval framework uses the year and multiplier information to convert the data to be stored in the datastore. All currency values are stored in base year currency units and all values are stored without exponents.
10.4.1.4 6.5. The Datastore
VisionEval changes the approach to storing model data from that of the GreenSTEP and RSPM models and related models. Those models stored data primarily in R data frames as binary files (rda files). The largest of these files are the simulated household files which store all of the information for all simulated households in an Azone (e.g. counties in GreenSTEP). All the data for households in the Azone are stored in a single data frame where each row corresponds to a record of an individual household and the columns are household attributes. Vehicle data for households are stored as lists in the data frame. This approach had some benefits:
- Storage and retrieval are part of the R language: one line of code to store a data frame, and one line of code to retrieve;
- It is easy to apply models to data frames; and
- Vehicle data can be stored as lists within a household data frame, eliminating the need to join tables.
The simplicity of this approach helped with getting GreenSTEP from a concept into an operational model quickly. However, several limitations have emerged as GreenSTEP and related models have been used in various applications including:
- Large amounts of computer memory are required when modeling Azones that have large populations. This necessitates either expanding computer memory or limiting the size of Azones;
- It is not easy to produce summary statistics from the simulated household files for a region; and
- The number of non-household data files has proliferated in order to store various aggregations for use in the model and for later summarization.
Finally, because the GreenSTEP/RSPM approach did not define a consistent data model, it does not sufficiently support the goal of modularity, and it does not support the use of alternative datastores. To overcome these limitations the VisionEval model system specifies a consistent datastore design. This design has been implemented in two types of datastores. One uses R binary files within a hierarchical directory structure. The other uses the HDF5 file format for storing model data. The HDF5 file format was developed by the National Center for Supercomputing Applications (NCSA) at the University of Illinois and other contributors to handle extremely large and complex data collections. For example, it is used to store data from particle simulations and climate models. It also is the basis for the new open matrix standard for transportation modeling, OMX.
VisionEval datastores are organized in a ‘column-oriented’ and hierarchical structure illustrated below. The lowest level of the hierarchy are datasets which are vectors of data values. This matches well the data objects (lists and data frames) commonly used in R programs and calculation methods which are commonly vectorized. Datasets in R datastores are R binary files. In HDF5 datastores, they are a portion of the HDF5 file and are called datasets in the HDF5 nomenclature. Datasets are organized in tables which are groups of datasets that all have the length. For example in the diagram below, Azone is a dataset containing the names of each of the Azones, and Age0to14 is a dataset containing the number of people of age 0 to 14 in each each Azone. These datasets and all those listed directly below them are contained in the Azone table. In R datastores, tables are represented by directories. In HDF5 nomenclature they are called groups. Tables that represent values for a particular model run year are grouped together. The year groups are named with the model run years (e.g. 2010, 2050) and they contain tables for every geographic level as well as ‘Household,’ ‘Worker,’ and ‘Vehicle’ tables. Tables that contain datasets whose values don’t vary by model run year are contained in the ‘Global’ group (directory). These include tables of model parameters, geographic tables, and any other table of datasets that apply to all model run years.
|____Global
| |____Model
| | ...
| |
| |____Azone
| | ...
| :
|
|____2010
| |____Region
| | ...
| |
| |____Azone
| | Azone
| | Marea
| | Age0to14
| | Age15to19
| | Age20to29
| | Age30to54
| | Age55to64
| | Age65Plus
| | ...
| |
| |____Bzone
| | ...
| |
| |____Marea
| | ...
| |
| |____Household
| | ...
| |
| |____Worker
| | ...
| |
| |____Vehicle
| | ...
| :
|
|
|____2050
| |____ ...
| | ...
: :
This structure is adequate to store all of the data that are used by the GreenSTEP/RSPM models and their offshoots. It can also be easily expanded to serve new modeling capabilites. For example if a module is added to model building stock, a ‘Buildings’ table could be added to each ‘forecast year’ group. In addition, this structure can accommodate matrix data as well as vector data, if a future module makes use of a distance matrix, that matrix could be added to either the ‘Global’ group or the ‘forecast years’ groups.
Note that the Azone table in the diagram includes an Marea dataset. This datasets identifies the Marea associated with each Azone. Every table includes datasets that enable data to be joined between tables. Some are geographic as in the example. Others use other identifiers. For example the Worker table includes a household identifier. Although the tables include the identifiers which enable data to be joined between tables, the VisionEval framework does not include special features for joining datasets that come from different tables. That is done by the module code.
10.4.1.5 6.6 The Model State File
The model state file, “ModelState.Rda,” maintains a record of all of the model run parameters and an inventory of the contents of the datastore. The software framework functions use this information to control the model run and to perform checks on module and data validity. The model state file contains a list which has the following components:
- Model: The name of the model
- Scenario: The name of the scenario
- Description: A description of the scenario
- Region: The name of the region being modeled
- BaseYear: The model base year
- Years: A list of years the model is being run for
- DatastoreName: The file name for the datastore
- DatastoreType: The datastore storage type. Currently implemented types are “RD” (R data) or “H5” (HDF5).
- Seed: The value to be used for the random seed
- RequiredVEPackages: The names of VisionEval module packages that are used in the model run
- LastChanged: The date and time of the last change to the model state
- Deflators: A data frame of deflator values by year
- Units: A data frame of default units for complex data types
- RequireVEPackages: A list of all the VisionEval packages that must be installed in order for the model to run
- LogFile: The file name of the log file
- Geo_df: A data frame containing the geographic definitions for the model
- BzoneSpecified: A logical value identifying whether Bzones are specified for the model
- CzoneSpecified: A logical value identifying whether Czones are specified for the model
- Datastore: A data frame containing an inventory of the contents of the datastore
- ModuleCalls_df: A data frame identifying the sequence of ‘runModule’ function calls and arguments
- ModulesByPackage_df: A data frame identifying the modules located in each of the VisionEval packages required to run the model
- DatasetsByPackage_df: A data frame identifying the datasets located in each of the VisionEval packages required to run the model
The Datastore component is updated every time the data is written to the datastore. This enables framework functions to ‘know’ the contents of the datastore without having to access the datastore. The Datastore component keeps track of all groups and datasets in the datastore and their attributes such as the length of tables and the specifications of datasets.
10.4.2 7. Overview of Module and Software Framework Layer Interactions
Modules are the heart of the VisionEval model system. Modules contain all of the code and parameters to implement submodels that are the building blocks of models. Modules are distributed in standard R packages. A VisionEval module package contains one or modules that implement related functions. For example, the VESimHouseholds package contains a set of modules related to creating a set of synthetic households and giving those households attributes including persons by age, workers by age, life cycle, and income. A module contains the following components:
- Documentation of the module and of submodel parameter estimation/calibration.
- Data and code for estimating the submodel(s) implemented by the module;
- Functions for estimating/calibrating parameters using regional data supplied by the user (if necessary);
- Data specifications for data that is to be loaded from input files, data that is to be loaded from the datastore, and data that is to be saved to the datastore; and,
- One or more functions for implementing the module.
The software framework provides all of the functionality for managing a model run. This includes:
- Checking module specifications for consistency with standards;
- Checking input files for compliance with module specifications;
- Processing input files to load the input data into the datastore;
- Simulating the data transactions in a model run to check whether the datastore will contains the data each module needs when the module needs it;
- Loading module packages;
- ‘Running’ modules in accordance with the ‘run_model.R’ script;
- Fetching from the datastore, data that is required by a module;
- Saving to the datastore, data that a module produces and specifies is to be saved; and,
- Converting measurement units and currency years when necessary.
When the software framework “runs” a module it does several things. First, it reads in the module data specifications and the main module function which performs the submodel calculations. Then it reads in all the datasets from the datastore that the module specifies. It also reads in the contents of the model state file. It puts these datasets into an input list and then it calls the main module function with this input list as the argument to the function call. This list, which by convention is called ‘L,’ contains 4 components: Global, Year, BaseYear, and G. The Global, Year, and BaseYear components contain lists which are composed of table components which are in turn composed of dataset components. The Global component relates to the ‘global’ group in the datastore. The Year component relates to the the group in the datastore for the model run year. For example, if the model run year is 2040, the Year component will contain data that is read from the ‘2040’ group in the datastore. The BaseYear component relates to the group in the datastore which represents the base year for the model. For example if the model base year is 2010, the BaseYear component will contain data that is read from the ‘2010’ group in the datastore. The list contains a BaseYear component as well as a Year component because modules may need to compute changes in a dataset between the base year and the model run year. Each of these components will contain a component for each table the the module requests data from. The table component is also a list which then contains components for all the requested datasets. For example, if a module needs the household income (‘Income’) and household size (‘HhSize’) datasets from the ‘Household’ table in the model run year (e.g. 2040), the Year component of ‘L’ will contain a ‘Household’ component which will contain an ‘Income’ component and a ‘HhSize’ component. The ‘G’ component of ‘L’ contains the model state list which contains all the information described in Section 6.6.
If a module calls another module, the list also contains a component named with the alias the module assigns to the called module. This component includes Global, Year, and BaseYear components containing datasets identified in the specifications for the called module. Module calling is explained in detail in section 8.1.2.
When the module executes, it returns a list containing all of the information that the module specifies is to be written to the datastore. By convention, this list is called ‘R.’ This list also has Global, Year, and BaseYear components which are structured in the same way that ‘L’ is structured. The table and dataset components of the list also include some attributes (metadata) which are described in Section 8.
10.5 Modules
All modules are made available in the form of standard R packages that meet framework specifications. Packages may be in source form or installed form. The following presentation refers to the source form of VisionEval packages. When a package is installed (built) the R system runs the scripts in the R directory and saves functions that are defined within and datasets that are saved by the script. The structure of this installed package is not covered by this document with the exception of the module_docs directory in the installed package. If the module developer follows the guidelines for module documentation, a module_docs directory will be included in the installed package. This will include module documentation in the form of markdown documents.
A package may contain more than one module. The package organization follows the standard organization of R packages. The structure is shown in the following diagram. The components are described below in the order that they are presented in the diagram. The file names are examples.
VESimHouseholds
| DESCRIPTION
| NAMESPACE
| LICENSE
|
|
|____R
| | CreateHouseholds.R
| | PredictWorkers.R
| | CreateEstimationDatasets.R
| | ...
|
|
|____inst
| | NOTICE
| |____extdata
| | pums_households.csv
| | pums_households.txt
| | pums_persons.csv
| | pums_persons.txt
| | ...
|
|
|____data-raw
| | some_processed_model_dataset.rda
| | ...
|
|
|____tests
|____scripts
| | test.R
| | test_functions.R
| | verspm_test.R
| | vestate_test.R
|
|____verspm
| | logs
| | | Log_CreateHouseholds.txt
| | | ...
| |
| | ModelState.Rda
|
|____vestate
| | ...
|
| ...
The DESCRIPTION and NAMESPACE files are standard files required by the R package system. There are good sources available for describing the required contents of these files (R Packages), so that will not be done here. Most of the entries in these files can be produced automatically from annotations in the R scripts that will be described next, using freely available tools such as devtools and RStudio. The LICENSE file contains the text for the Apache 2 license which is the open source license that should be used for VisionEval packages. The accompanying Apache 2 NOTICE file is in the inst directory.
10.5.0.1 8.1. The R Directory
The R directory is where all the R scripts are placed which define the modules that are included in the package. Each module is defined by a single R script which has the name of the module (and the .R file extension). A module script does 4 things:
1) It specifies the model and estimates model parameters. Model estimation datasets may be processed by the script or may be loaded from other sources (see below).
2) It defines all the specifications for data that the module depends on.
3) It defines all of the functions that implement the submodel.
4) It includes test code used during module development
When a binary (installed) package is built from the source package, each R script is run; estimation datasets are processed, model parameters are saved, module specifications are saved, functions are parsed into a more efficient form. Following section describe the structure of a module R script in more detail. An example of the AssignTransitService module script from the VETransportSupply package is included in Appendix E.
By convention, the module script is organized into 4 sections reflecting the list above. Following sections 8.1.1 to 8.1.4 explain each module script section in detail. Section 8.1.1 explains how model specification and estimation is handled. Subsection 8.1.1.1 focuses in on the handling of model estimation data which, in some cases, may use specialized scripts. Section 8.1.2 explains how to write module specifications which tell the framework how the module is to be run, what input data are to be processed, what data are to be retrieved from the datastore, and what data are to be saved to the datastore. Subsection 8.1.2.1 focuses in on the OPTIONAL specification that module developers can use to enable optional model inputs. Subsection 8.1.2.2 focues in the the CALL specification that enables modules to call other modules to perform calculation services. Section 8.1.3 describes how to write a main function and supporting functions to implement a module. Section 8.1.4 explains test code that is commonly included in a module to assist the module developer in the module development process. Finally, Section 8.1.5 explains a special module named Initialize that may be included in a package. The purpose of this module is to enable module developers to include specialized input data checks and preparation.
10.5.0.1.1 8.1.1 Model Specification and Estimation
As the name suggests, this section of the script handles tasks related to processing model estimation data, model specification, and model parameter estimation. This should be done in a manner which enables others to understand the structure of the model, the data it is based on, and how parameters are estimated. It should enable others to replicate the process and to test modifications. Typically, model specification and estimation code does 4 things: loading the data to be used, defining functions that carry out key specification and estimation procedures, applying the functions to produce objects that define the model(s), and saving the model objects.
Models vary in their complexity. In some modules the model may simply be an algorithm with decision rules or asserted parameters. In this case, there is no need to fill out this section of the script. For example, the AssignLifeCycle module in the VESimHouseholds package assigns lifecycle codes to households using rules regarding the numbers of adults, children, and workers in the household. In cases like this, the model estimation and specification section will be empty. In other cases, models will be quite complex and this section of the script will be extensive. For example, the “CalculateHouseholdDvmt” module does the following in the process of building models and comprises hundreds of lines of code:
- Estimate a binomial logit model of the probability that a household has any DVMT
- Estimate a linear model of the amount of household DVMT given that a household has any DVMT
- Estimate a dispersion factor for the linear model so that the variance of modeled DVMT matches the observed variance
- Simulate household DVMT 1000 times by stochastically applying the binomial and linear models
- Calculate quantiles of simulated DVMT
- Estimate a linear model of average DVMT using the simulation data
- Estimate models of the DVMT quantiles as a function of the average DVMT
Model estimation data preparation may also be extensive. Whether it is or not, documenting the data used in building a model and making those data available are key to making the models reproducible and extensible. Moreover, for some modules it is important that model users be able to have model parameters reflect the conditions for their region. For example, the CreateHouseholds, PredictWorkers, and PredictIncome modules use Census public use microdata sample data for the region to estimate parameters. There are several ways that model estimation data can be handled in the VisionEval model system. These are described in detail in Section 8.1.1.1.
The model estimation code should also save objects to use in documenting the modules such as: 1. Model summary statistics such as is produced when a model object (such as a linear model estimated using the lm function) is processed by the summary function; 2. Data frames, tables, and matrices; and, 3. Graphs or other static data visualizations.
The first two should be saved as objects just like any other object saved to implement a model. It is suggested that these documentation objects be stored in a list along with the model that they document. For example, follow is a portion of the PredictIncome.R script which saves (and documents) a list which contains the household income model including documentation of the summary statistics for the model.
#Save the household income model
#-------------------------------
#' Household income model
#'
#' A list containing the income model equation and other information needed to
#' implement the household income model.
#'
#' @format A list having the following components:
#' \describe{
#' \item{Type}{a string identifying the type of model ("linear")}
#' \item{Formula}{makeModelFormulaString(IncModel_LM)}
#' \item{PrepFun}{a function that prepares inputs to be applied in the linear model}
#' \item{OutFun}{a function that transforms the result of applying the linear model}
#' \item{Summary}{the summary of the linear model estimation results}
#' \item{SearchRange}{a two-element vector specifying the range of search values}
#' }
#' @source PredictIncome.R script.
"HHIncModel_ls"
usethis::use_data(HHIncModel_ls, overwrite = TRUE)The Summary component of this list shows the summary statistics for the model. As explained in the module documentation section below, this can be automatically inserted into the module documentation. It is recommended that the capture.output function be used rather than the print function to capture model summary statistics because the print function will insert line numbers. This method can be used to save other text that can then be inserted into module documentation. This example also shows how objects are documented and saved.
It can also be useful to save data frames, tables, matrices to use in the model documentation. These can be saved like any other data object and then inserted as described in the module documentation section.
Graphs or other visualizations are saved in a different manner. If these are saved as an image file in “png” format, they can be inserted into the module documentation. They must be saved to the “data” directory to do so. Following is an example:
#Plot comparison of observed and estimated income distributions
png(
filename = "data/reg-hh-inc_obs-vs-est_distributions.png",
width = 480,
height = 480
)
plot(
density(IncObs_),
xlim = c(0, 200000),
xlab = "Annual Dollars ($2000)",
main = "Distributions of Observed and Predicted Household Income \nRegular Households"
)
lines(density(IncEst_), lty = 2)
legend("topright", legend = c("Observed", "Predicted"), lty = c(1,2))
dev.off()There are many ways that a module developer can code the model specification and parameter estimation procedures. However this is done, the code should be well organized and commented so that it is understandable to reviewers. All code should follow the VisionEval coding guidelines. In addition, it is highly recommended that code be grouped into functions to aid understandability and reduce unnecessary code repetition which can lead to errors. Complex functions should be well documented. Function documentation is done using Roxygen syntax. Following is an example of the code which estimates a housing choice model (single-family vs. multifamily) in the PredictHousing module of the VELandUse package.
#Define a function to estimate housing choice model
#--------------------------------------------------
#' Estimate housing choice model
#'
#' \code{estimateHousingModel} estimates a binomial logit model for choosing
#' between single family and multifamily housing
#'
#' This function estimates a binomial logit model for predicting housing choice
#' (single family or multifamily) as a function of the supply of housing of
#' these types and the demographic and income characteristics of the household.
#'
#' @param Data_df A data frame containing estimation data.
#' @param StartTerms_ A character vector of the terms of the model to be
#' tested in the model.
#' @return A list which has the following components:
#' Type: a string identifying the type of model ("binomial"),
#' Formula: a string representation of the model equation,
#' PrepFun: a function that prepares inputs to be applied in the binomial model,
#' OutFun: a function that transforms the result of applying the binomial model.
#' Summary: the summary of the binomial model estimation results.
#' @import visioneval stats
#Define function to estimate the income model
estimateHousingModel <- function(Data_df, StartTerms_) {
#Define function to prepare inputs for estimating model
prepIndepVar <-
function(In_df) {
Ah <-
c("Age15to19",
"Age20to29",
"Age30to54",
"Age55to64",
"Age65Plus")
Out_df <-
data.frame(t(apply(In_df[, Ah], 1, function(x) {
AgeLvl_ <- 1:5 #Age levels
HhAgeLvl_ <- rep(AgeLvl_, x)
HeadOfHh_ <- numeric(5)
if (max(HhAgeLvl_) < 5) {
HeadOfHh_[max(HhAgeLvl_)] <- 1
} else {
if (all(HhAgeLvl_ == 5)) {
HeadOfHh_[5] <- 1
} else {
NumMidAge <- sum(HhAgeLvl_ %in% c(3, 4))
NumElderly <- sum(HhAgeLvl_ == 5)
if (NumMidAge > NumElderly) {
HeadOfHh_[max(HhAgeLvl_[HhAgeLvl_ < 5])] <- 1
} else {
HeadOfHh_[5] <- 1
}
}
}
HeadOfHh_
})))
names(Out_df) <- paste0("Head", Ah)
Out_df$HhSize <- In_df$HhSize
Out_df$Income <- In_df$Income
Out_df$RelLogIncome <- log1p(In_df$Income) / mean(log1p(In_df$Income))
Out_df$Intercept <- 1
Out_df
}
EstData_df <- prepIndepVar(Data_df)
EstData_df$SingleFamily <- as.numeric(Data_df$HouseType == "SF")
#Define function to make the model formula
makeFormula <-
function(StartTerms_) {
FormulaString <-
paste("SingleFamily ~ ", paste(StartTerms_, collapse = "+"))
as.formula(FormulaString)
}
#Estimate model
HouseTypeModel <-
glm(makeFormula(StartTerms_), family = binomial, data = EstData_df)
#Return model
list(
Type = "binomial",
Formula = makeModelFormulaString(HouseTypeModel),
Choices = c("SF", "MF"),
PrepFun = prepIndepVar,
Summary = summary(HouseTypeModel)
)
}
#Estimate the binomial logit model
#---------------------------------
#Load the household estimation data
Hh_df <- VESimHouseholds::Hh_df
#Select regular households
Hh_df <- Hh_df[Hh_df$HhType == "Reg",]
Hh_df$Income[Hh_df$Income == 0] <- 1
#Estimate the housing model
HouseTypeModelTerms_ <-
c(
"HeadAge20to29",
"HeadAge30to54",
"HeadAge55to64",
"HeadAge65Plus",
"RelLogIncome",
"HhSize",
"RelLogIncome:HhSize"
)
HouseTypeModel_ls <- estimateHousingModel(Hh_df, HouseTypeModelTerms_)
rm(HouseTypeModelTerms_)As can be seen in this example, most of the code defines and documents a function which estimates a binomial choice model given a estimation dataset and a list of model variable names. This function does a number of things including transforming estimation data, creating a model formulation from the list of independent variables, estimating the variables, and returning a list of model components. After the function is defined, it is applied using the model estimation data and a specified list of independent variables. This approach to model specification and estimation makes it easy for a module developer to try out different model specifications, measuring their relative performance, and documenting the results.
The approach used in this example takes advantage of some helper functions in the framework software. The makeModelFormulaString function takes an R model formula object and converts it into a string representation. When the model is applied in the module, the string is parsed and evaluated with a data frame of independent variables. This provides a compact way to store a model and a fast way to apply it. Several other framework functions assist with applying models in this form. The applyBinomialModel applies a binomial logit model. The applyLinearModel applies a linear regression model. Both of these functions enable the models to self calibrate to match an input target. For example, the PredictHousing module adjusts the constant of the model so that the proportions of households in single-family vs. multifamily dwelling units matches the housing supply in the area. The applyBinomialModel does this efficiently by calling the binarySearch function which implements a binary search algorithm.
Module developers are not limited to using the previous approach to specifying and implementing a model. Most if not all R functions that estimate models also have companion prediction functions to apply the estimated model to a new set of data. This is often the simplest and best way to apply a model. One drawback of this approach, however, is that the object that is returned by the model estimation function and that is used in making a prediction is often very large because it includes a copy of all of the estimation data and datasets used to calculate model statistics. If the model estimation datasets are very large, this will pose a problem for keeping the module package in a central repository. In that case, the module code needs to remove parts of the model object that are not necessary for prediction. For example, the AssignVehicleOwnership module in the VEHouseholdVehicles package uses an ordered logit model to predict household auto ownership for households owning one or more vehicles. The model is estimated using data from the 2001 NHTS so the model object returned by the estimation function (clm in the ordinal package) is large. Portions of the model object not needed by the corresponding prediction function are removed. It is often a matter of trial and error to find out how much can be removed from the model object without adversely affecting the prediction function.
#Model number of vehicles of non-zero vehicle households
EstData_df <- EstData_df[EstData_df$ZeroVeh == 0,]
EstData_df$VehOrd <- EstData_df$NumVeh
EstData_df$VehOrd[EstData_df$VehOrd > 6] <- 6
EstData_df$VehOrd <- ordered(EstData_df$VehOrd)
AutoOwnModels_ls$Metro$Count <-
clm(
VehOrd ~ Workers + LogIncome + DrvAgePop + HhSize + OnlyElderly + IsSF +
IsUrbanMixNbrhd + LogDensity + TranRevMiPC,
data = EstData_df,
threshold = "equidistant"
)
#Trim down model
AutoOwnModels_ls$Metro$Count[c("fitted.values", "model", "y")] <- NULLAfter a model has been estimated, the model objects which embody the model need to be saved as part of the package. There is a standard way of doing this which is illustrated in the following example of saving the model in the PredictHousing module. The housing prediction model object is a list called ‘HouseTypeModel_ls.’ This list is saved in the ‘data’ directory of the package. By saving the model object in the ‘data’ directory, it can be addressed directly in the module function that applies the model. Note how documentation is handled.
#Save the housing choice model
#-----------------------------
#' Housing choice model
#'
#' A list containing the housing choice model equation and other information
#' needed to implement the housing choice model.
#'
#' @format A list having the following components:
#' \describe{
#' \item{Type}{a string identifying the type of model ("binomial")}
#' \item{Formula}{makeModelFormulaString(HouseTypeModel)}
#' \item{PrepFun}{a function that prepares inputs to be applied in the model}
#' \item{Summary}{the summary of the binomial logit model estimation results}
#' \item{SearchRange}{a two-element vector specifying the range of search values}
#' }
#' @source PredictHousing.R script.
"HouseTypeModel_ls"
usethis::use_data(HouseTypeModel_ls, overwrite = TRUE)
The use_data function in the usethis package saves the data and documention correctly in the package and simplifies the procedure for doing so.
Although model estimation code is usually included in the module script, in some instances it may be more understandable to estimate the models for several modules in one script. This is what is done in the CreateSimBzoneModels.R script in the VESimLandUse package. This script estimates all the models for modules in the package. Model estimation is handled this way because the same model estimation dataset is used for all the models and the models build upon each other. Including all the model estimations in one file makes it easier to code and easier to review.
10.5.0.1.1.1 8.1.1.1 Model Estimation Datasets
Model estimation datasets may be read in from several sources. If model estimation data are large and are used by multiple modules located in different packages, then they may be housed in their own package. This is the case with the 2001 NHTS data which are in the VE2001NHTS package. The estimation data in the package can be directly addressed in the model estimation code using the standard ‘PackageName::DatasetName’ notation as shown in the following example:
Hh_df <- VE2001NHTS::Hh_dfIf estimation data are retrieved from another package as in this case, the DESCRIPTION file for the package must list the package from which the data are retrieved in the Imports section. For example, the Imports section of the VEHouseholdTravel package reads as follows. The VE2001NHTS package is listed as are other packages that this package relies on.
Imports:
visioneval,
devtools,
VE2001NHTS,
data.tableIf estimation data is to be shared among several modules that are all within the same package, then all the data preparation can be done in one script which is run first when the package is built. This approach is used in the VESimHouseholds package. The CreateEstimationDatasets.R script loads and processes a Census public use microdata sample dataset that is used in estimating models in 3 modules that are in the package (CreateHouseholds, PredictWorkers, PredictIncome). When, as in this case, a separate script is used to prepare the estimation data for several modules in the package, the data are saved in the data directory of the package and the other scripts load the dataset from that directory. The following extracts shows how a processed PUMS household dataset (a data frame named Hh_df) is saved in the CreateEstimationDatasets.R script. The lines starting with #' are data documentation in roxygen2 form. Although data documentation can be tedious, especially for large datasets, it is should always be done and is an important aspect of the VisionEval model system.
#' Household data from Census PUMS
#'
#' A household dataset containing the data used for estimating the
#' CreateHouseholds, PredictWorkers, PredictLifeCycle, PredictIncome, and
#' PredictHouseType modules derived from from year 2000 PUMS data for Oregon.
#'
#' @format A data frame with 65988 rows and 17 variables (there may be a
#' different number of rows if PUMS datasets are used for different areas):
#' \describe{
#' \item{Age0to14}{number of persons in 0 to 14 age group}
#' \item{Age15to19}{number of persons in 15 to 19 age group}
#' \item{Age20to29}{number of persons in 20 to 29 age group}
#' \item{Age30to54}{number of persons in 30 to 54 age group}
#' \item{Age55to64}{number of persons in 55 to 64 age group}
#' \item{Age65Plus}{number of persons 65 years or older}
#' \item{Wkr15to19}{number of workers in 15 to 19 age group}
#' \item{Wkr20to29}{number of workers in 20 to 29 age group}
#' \item{Wkr30to54}{number of workers in 30 to 54 age group}
#' \item{Wkr55to64}{number of workers in 55 to 64 age group}
#' \item{Wkr65Plus}{number of workers 65 years or older}
#' \item{AvePerCapInc}{average per capita income of PUMA, nominal $}
#' \item{HouseType}{housing type (SF = single family, MF = multifamily)}
#' \item{Income}{annual household income, nominal 1999$}
#' \item{HhSize}{number of persons in household}
#' \item{HhType}{household type (Reg = regular household, Grp = group quarters)}
#' \item{HhWeight}{household sample weight}
#' }
#' @source CreateEstimationDatasets.R script.
"Hh_df"
devtools::use_data(Hh_df, overwrite = TRUE)
rm(Hh_df)If as in this case, one script processes the estimation data for use by several modules, it is important that the data processing script be run prior to the other scripts when the package is built. Otherwise the processed estimation data will not be available to the module packages that need it. Managing the order of package script processing is handled by the Collate section of the package DESCRIPTION file. Following is an example from the VESimHouseholds package:
Collate:
'CreateEstimationDatasets.R'
'CreateHouseholds.R'
'PredictWorkers.R'
'PredictIncome.R'
'AssignLifeCycle.R'The best way to establish the proper collation (i.e. ordering) of script execution is to carry out model estimation through the definition and invocation of a function, and to have an 3 statement which lists the estimation data processing script in the function documentation using Roxygen syntax. When the package documentation is compiled, the Collate section of the package DESCRIPTION file will be filled out properly. Following is an example of function documentation containing a 4 statement from the CreateHouseholds.R module script in the VESimHouseholds package:
#Define a function to estimate household size proportion parameters
#------------------------------------------------------------------
#' Calculate proportions of households by household size
#'
#' \code{calcHhAgeTypes} creates a matrix of household types and age
#' probabilities.
#'
#' This function produces a matrix of probabilities that a person in one of six
#' age groups is in one of many household types where each household type is
#' determined by the number of persons in each age category.
#'
#' @param HhData_df A dataframe of household estimation data as produced by the
#' CreateEstimationDatasets.R script.
#' @param Threshold A number between 0 and 1 identifying the percentile
#' cutoff for determining the most prevalent households.
#' @return A matrix where the rows are the household types and the columns are
#' the age categories and the values are the number of persons.
#' @include CreateEstimationDatasets.R
#' @exportIf model estimation data are only used by a single module, then the processing code should be included in the module script.
There are several places where model estimation datasets may be located depending on the size of the datasets, whether the datasets were preprocessed using other tools or scripts, and whether the intention of the module developer is to enable model users to customize model estimation to reflect data for the region where the model is to be applied.
For some modules, it is desirable that model parameters be estimated to reflect data for the region where the model is to be applied. This is the case for the CreateHouseholds, PredictWorkers, and PredictIncome modules in the VESimHouseholds package because household and worker age compositions and income distributions are likely to vary by region. For modules like these, the estimation data should be included in the ‘inst/extdata’ directory of the package in the form of CSV-formatted text files. The data files should be accompanied by text files having the same names but with a ‘.txt’ extension rather than a ‘.csv’ extension. The purpose of the text files is to document the corresponding data files so that model users will know how to obtain and prepare datasets for their region that can be used in model estimation. For example, the ‘inst/extdata’ directory of the VESimHouseholds package contains the following files: ‘pums_households.csv,’ ‘pums_households.txt,’ ‘pums_persons.csv,’ and ‘pums_persons.txt.’
If a module’s estimation datasets are to be loaded from ‘csv’ files in the ‘inst/extdata’ directory, then the module code should include procedures to check those data for correctness to assure that the model parameters will be estimated correctly. The framework includes as function, processEstimationInputs, to assist with this task. The processEstimationInputs function has 3 arguments: Inp_ls, FileName, and ModuleName. ModuleName is the name of the module that is doing the processing. The sole purpose of this argument is to identify the module in any error messages that are written to the log. FileName is the name of the file in the ‘inst/extdata’ directory to be processed. Inp_ls is a list of data specifications that are used in checking the correctness of the data. This is explained in more detail below. The processEstimationInputs function returns a data frame which contains the data in the specified file, if those data have no errors. Following is an example of how the processEstimationInputs function is used:
Hh_df <- processEstimationInputs(
Inp_ls = PumsHhInp_ls,
FileName = "pums_households.csv",
ModuleName = "CreateEstimationDatasets")Before the estimation data may be checked, the script must describe specifications for the data. The structure of the specifications is best described using an example. The following code snippet is from the CreateEstimationDatasets.R script:
PumsHhInp_ls <- items(
item(
NAME =
items("SERIALNO",
"PUMA5",
"HWEIGHT",
"UNITTYPE",
"PERSONS"),
TYPE = "integer",
PROHIBIT = c("NA", "< 0"),
ISELEMENTOF = "",
UNLIKELY = "",
TOTAL = ""
),
item(
NAME = "BLDGSZ",
TYPE = "integer",
PROHIBIT = c("< 0"),
ISELEMENTOF = "",
UNLIKELY = "",
TOTAL = ""
),
item(
NAME = "HINC",
TYPE = "double",
PROHIBIT = c("NA"),
ISELEMENTOF = "",
UNLIKELY = "",
TOTAL = ""
)
)The meanings of these specifications are as follows:
- NAME This is the name(s) of the data column in the file. The name must be a character string (i.e. surrounded by quotation marks). If multiple columns of the file have the same specifications except for their names, they can listed as in the first item in the example. This method avoids a lot of redundant specifications. Note that the order of specifications does not need to be the same as the order of the columns in the file. Also note that it is OK if the file contains columns that are not specified, as long as it contains all of the columns that are specified. Columns that are not listed are ignored.
- TYPE This the data type of the data contained in the column. Allowable types are the 4 primitive types recognized by the R language (integer, double, character, and logical), the complex types listed in section 6.4, or ‘compound.’ The type must be a character string.
- PROHIBIT This is a character vector which identifies all prohibited data conditions. For example, the specification for the “PERSONS” data column in the example above is c(“NA,” “< 0”). This means that there cannot be any values that are undefined (NA) or less than 0. The symbols that may be used in a PROHIBIT specification are: NA, ==, !=, <, <=, >, >= (i.e. undefined, equal to, not equal to, less than, less than or equal to, greater than, greater than or equal to). Note that prohibited conditions must be represented as character strings. If there is more than one prohibited condition, as in the example, the conditions must be entered as an R vector using the ‘c()’ function, as shown in the example. The absence of prohibited conditions is represented by an empty character string (i.e. ““).
- ISELEMENTOF This is a vector which specifies the set of allowed values. It is used when the input values must be elements of a set of discrete values. The vector describing the set must be of the same type as is specified for the input data. Vectors of elements are entered using the ‘c()’ function. For example, if the entries in a column of data could only have the values ‘urban’ or ‘rural,’ then the specification owould be written as c(”urban”, “rural”). The absence of a specification for this is represented by an empty character string.
- UNLIKELY This is a vector of conditions that while not prohibited, are not likely to occur. While conditions identified in the PROHIBIT and ISELEMENTOF specifications will produce an error if they are not met (thereby stopping the calculation of parameters), the conditions identified in the UNLIKELY specification will only produce a warning message. Vectors of conditions are entered using the ‘c()’ function.
- TOTAL This specifies a required total value for the column of data. This is useful if the data represents proportions or percentages and must add up to 1 or 100. The absence of a specification for this is represented by an empty character string.
In some instances it is impractical to include the model estimation data as files in the ‘inst/extdata’ directory, and not necessary for model users to provide regional model estimation data. If, for example, the source data has confidential elements, it may be necessary to preprocess the data to anonymize it before including in the package; or the source data may be too large to include as a text file in the package. In these cases, source data may be processed outside of the package and then the processed datasets included in the package as datasets in R binary files. If that is done, the binary data files should be placed in a directory named ‘data-raw’ in the package.
For large datasets, it is also possible to have them stored remotely and to have the module code retrieve them from remote storage. This is what is done in the Make2001NHTSDataset.r script in the VE2001NHTS package. Zip archives of the public use datasets for the 2001 National Household Travel Survey are stored in a GitHub repository (https://github.com/VisionEval/NHTS2001). There are 4 zip archive files stored in the ‘data’ directory of the repository: DAYPUB.zip, HHPUB.zip, PERPUB.zip, and VEHPUB.zip. Each zip archive contains a compressed ‘csv’ formatted text file containing the data of interest. Because the process of downloading and unzipping some of the files takes an appreciable amount of time, the script checks whether that has already been done (by the presence of a file in the ‘data-raw’ directory). If it has not been done, the script calls a function which handles the downloading, unzipping, and reading of the file. It also cleans up temporary files created when the zip archive was downloaded and unzipped.
If a module is to download datafiles from remote storage, the module developer will need to write R code to handle the requisite tasks. The framework does not build in any functionality to do this. The coding is not complicated as the following example of commented snippets of code from the Make2001NHTSDataset.r script illustrate. The first section of code identifies the address of the data repository. The second section defines a function which handles downloading, unzipping, and reading a dataset stored in the repository, and cleaning up temporary files created in the process. The third section applies the handler function to download NHTS public use household data (HHPUB), selects relevant data fields, and saves those data as an R binary file to the ‘data-raw’ directory. If the R binary file already exists, the data are not downloaded from the repository but are loaded from the saved file instead.
#Identify the code repository. Note that in order to access data
#files from a GitHub repository, the beginning of the address must
#be 'raw.githubusercontent.com' not 'github.com'
Nhts2001Repo <-
"https://github.com/VisionEval/NHTS2001/blob/master/data"
#Define a function to handle retrieving a zipped dataset from the
#repository, unzipping it, reading the unzipped file, and cleaning
#up temporary files
getZipDatasetFromRepo <- function(Repo, DatasetName) {
ZipArchiveFileName <- paste0(DatasetName, ".zip")
CsvFileName <- paste0(DatasetName, ".csv")
download.file(paste0(file.path(Repo, ZipArchiveFileName),"?raw=true"), ZipArchiveFileName)
Data_df <- read.csv(unzip(ZipArchiveFileName), as.is = TRUE)
file.remove(ZipArchiveFileName, CsvFileName)
Data_df
}
#Download NHTS 2001 public use household data from repository and
#process if it has not already been done
if (!file.exists("data-raw/Hh_df.rda")) {
Hh_df <- getZipDatasetFromRepo(Nhts2001Repo, "HHPUB")
Keep_ <- c("HOUSEID", "AGE_P1", "AGE_P2", "AGE_P3", "AGE_P4", "AGE_P5", "AGE_P6",
"AGE_P7", "AGE_P8", "AGE_P9", "AGE_P10", "AGE_P11", "AGE_P12", "AGE_P13",
"AGE_P14", "CENSUS_D", "CENSUS_R", "DRVRCNT", "EXPFLHHN", "EXPFLLHH",
"FLGFINCM", "HBHRESDN", "HBHUR", "HBPPOPDN", "HHC_MSA", "HHFAMINC",
"HHINCTTL", "HHNUMBIK", "HHR_AGE", "HHR_DRVR", "HHR_RACE", "HHR_SEX",
"HHSIZE", "HHVEHCNT", "HOMETYPE", "HTEEMPDN", "HTHRESDN", "HTHUR",
"HTPPOPDN", "LIF_CYC", "MSACAT", "MSASIZE", "RAIL", "RATIO16V",
"URBAN", "URBRUR", "WRKCOUNT", "CNTTDHH")
Hh_df <- Hh_df[, Keep_]
save(Hh_df, file = "data-raw/Hh_df.rda", compress = TRUE)
} else {
load("data-raw/Hh_df.rda")
}10.5.0.1.2 8.1.2 Module Specifications
The module specifications section of the module script provides specifications that are used by the framework in a number of ways. The specifications:
- Identify the level of geography that the model is to be run at;
- Describe all the data to be loaded from input files and enable unit conversions;
- Describe all the data that the module needs from the datastore in order to run;
- Describe all the data that the module produces that needs to be saved to the datastore;
- Identify other modules the module needs to call in order to do its calculations;
- Allow the framework to check and load all input files before any modules are run;
- Allow the framework to check that modules are compatible with one another in terms of the data that they produce and consume;
- Allow the sequence of model steps (module calls) to be simulated to check whether each module will have the data it needs when it is run; and,
- Provide detailed documentation about all data that is produced by a model run.
These specifications are declared in a list that is similar to the list for specifying model estimation data described in Section 8.1.1.1. A full example is shown in Appendix E. Following is a skeleton of a module specifications list for a module whose name is MyModule. The name of the specifications list must be the concatenation of the module name and Specifications, so in this example the name is MyModuleSpecifications. Note that the functions items and item are aliases for the R list function. Their purpose is to make the specifications easier to read.
MyModuleSpecifications <- list(
RunBy = ...,
NewInpTable = items(
item(
...
),
item(
...
)
),
NewSetTable = items(
item(
...
),
item(
...
)
),
Inp = items(
item(
...
),
item(
...
)
),
Get = items(
item(
...
),
item(
...
)
),
Set = items(
item(
...
),
item(
...
)
),
Call = items(
item(
...
),
item(
...
)
)
)Following are detailed descriptions and examples of each component of the specifications list.
The RunBy component specifies the level of geography that the model is to be run at. For example, the congestion submodel in the GreenSTEP and RSPM models runs at the Marea level. This specification is used by the software framework to determine how to index data that is read from the datastore and data that is written to the datastore. Acceptable values are “Region,” “Azone,” “Bzone,” and “Marea.” The RunBy specification looks like the following example:
RunBy = "Marea",The NewInpTable and NewSetTable components specify any new tables that need to be created in the datastore to accommodate input data or data produced by a module respectively. The following specifications are required for each new table that is to be created.
- TABLE: the name of the table that is to be created; and,
- GROUP: the type of group the table is to be put into. There are 3 group types: Global, BaseYear, and Year. If Global, the table is created in the global group of the datastore. If BaseYear the table is created in the year group for the base year and only in that year group. For example, if the model base year is 2010, the table will be created in the 2010 group. If Year, the table will be created in the group for every model run year. For example, if the run years are 2010 and 2040, the table will be created in both the 2010 group and the 2040 group.
Following is an example of a NewSetTable specification for creating a Vehicle table in each model run year group.
NewSetTable = items(
item(
TABLE = "Vehicle",
GROUP = "Year"
)
),The Inp component specifies all model inputs that the module requires. Each item in the list describes one or more data fields in an input file. Each item must have the following attributes (except for the OPTIONAL attribute):
- NAME: the names of one or more data fields in the input file, and the names used for the datasets when they are loaded into the datastore. The names must match corresponding column names with the exception that column names in input files may contain year and multiplier modifiers as described in Section 6.4.;
- FILE: the name of the file that contains the data fields;
- TABLE: the name of the datastore table the datasets will be put into;
- GROUP: the type of group where the table is located in the datastore (i.e. Global, Year, BaseYear);
- TYPE: the data type (e.g. double, distance, compound);
- UNITS: the measurement units for the data;
- NAVALUE: the value used to represent NA (i.e. missing value) in the datastore;
- SIZE: the maximum number of characters for character data (or 0 for numeric data);
- PROHIBIT: data values that are prohibited or “” if not applicable (see Section 8.1.1.1 for explanation);
- ISELEMENTOF: allowed categorical data values or “” if not applicable (see Section 8.1.1.1 for explanation);
- UNLIKELY: data conditions that are unlikely or “” if not applicable (see Section 8.1.1.1 for explanation);
- TOTAL: the total for all values (e.g. 1) or “” if not applicable;
- DESCRIPTION: descriptions of the data corresponding to the names in the NAME attribute; and,
- OPTIONAL: optional specification which identifies whether dataset is optional (see section 8.1.2.1)
Following is an example of the Inp component for the PredictHousingSpecifications in the VELandUse package. The Inp specifications include two items. Each item lists one or more field names (i.e. column names in the input file) in the NAME attribute. Multiple field names can be listed in an item if all the other attributes except for DESCRIPTION are the same for all the fields. The descriptions in the DESCRIPTION attribute must correspond in order to the dataset names in the NAME attribute in order for them to be stored correctly in the datastore. The values that may be entered for the TYPE and UNITS attributes are described in Section 6.3. It should be noted that the UNITS attribute must not include ‘year’ (for ‘currency’ type) or ‘multiplier’ information. That information is part of the input file field names instead (where relevant). This is explained in more detail in Section 6.4. The framework uses the information in the Inp to read the input files, check whether the data are correct, and save the data to the correct location in the datastore.
Inp = items(
item(
NAME =
items(
"SFDU",
"MFDU",
"GQDU"),
FILE = "bzone_dwelling_units.csv",
TABLE = "Bzone",
GROUP = "Year",
TYPE = "integer",
UNITS = "DU",
NAVALUE = -1,
SIZE = 0,
PROHIBIT = c("NA", "< 0"),
ISELEMENTOF = "",
UNLIKELY = "",
TOTAL = "",
DESCRIPTION =
items(
"Number of single family dwelling units (PUMS codes 01 - 03) in zone",
"Number of multi-family dwelling units (PUMS codes 04 - 09) in zone",
"Number of qroup quarters population accommodations in zone"
)
),
item(
NAME = items(
"HhPropIncQ1",
"HhPropIncQ2",
"HhPropIncQ3",
"HhPropIncQ4"),
FILE = "bzone_hh_inc_qrtl_prop.csv",
TABLE = "Bzone",
GROUP = "Year",
TYPE = "double",
UNITS = "NA",
NAVALUE = -1,
SIZE = 0,
PROHIBIT = c("NA", "< 0"),
ISELEMENTOF = "",
UNLIKELY = "",
TOTAL = "",
DESCRIPTION =
items(
"Proportion of Bzone households (non-group quarters) in 1st quartile of Azone household income",
"Proportion of Bzone households (non-group quarters) in 2nd quartile of Azone household income",
"Proportion of Bzone households (non-group quarters) in 3rd quartile of Azone household income",
"Proportion of Bzone households (non-group quarters) in 4th quartile of Azone household income"
)
)
),The Get component contains one or more items that identify data that the module need to have retrieved from the datastore. Note that the Get component must identify all datasets the module requires (other than those that are included in the package), including those specified in the Inp component. The datasets identified in the Inp component are not automatically made available to the module. Each item in the Get component specifies attributes for one or more related datasets as follows:
- NAME: the names of one or more datasets to be loaded;
- TABLE: the name of the table that the datasets are located in;
- GROUP: the type of group where the table is located in the datastore (i.e. Global, Year, BaseYear);
- TYPE: the data type (e.g. double, distance, compound);
- UNITS: the measurement units for the data;
- PROHIBIT: data values that are prohibited or “” if not applicable (see Section 8.1.1.1 for explanation);
- ISELEMENTOF: allowed categorical data values or “” if not applicable (see Section 8.1.1.1 for explanation); and,
- OPTIONAL: optional specification which identifies whether dataset is optional (see section 8.1.2.1)
Unlike the ‘Inp’ specifications, the ‘Get’ specifications for UNITS must include ‘year’ information for ‘currency’ types. This is necessary in order for the framework to convert the currency data being requested from the datastore to the year denomination that the module needs. The UNITS attribute may also include a multiplier specification if the module needs the values to be modified in that way. For example, if the module needs VMT in thousands of miles, the UNITS specification would be ‘MI.1e3.’ Section 6.3 provides more information on how ‘year’ and ‘multiplier’ options are added to a UNITS specification. Following is an example of the Get component of the AssignRoadMilesSpecifications in the VETransportSupply package.
Get = items(
item(
NAME = "Marea",
TABLE = "Marea",
GROUP = "Year",
TYPE = "character",
UNITS = "ID",
PROHIBIT = "",
ISELEMENTOF = ""
),
item(
NAME =
items(
"FwyLaneMi",
"ArtLaneMi"),
TABLE = "Marea",
GROUP = "Year",
TYPE = "distance",
UNITS = "MI",
PROHIBIT = c("NA", "< 0"),
ISELEMENTOF = ""
),
item(
NAME = "Marea",
TABLE = "Bzone",
GROUP = "Year",
TYPE = "character",
UNITS = "ID",
PROHIBIT = "",
ISELEMENTOF = ""
),
item(
NAME = "UrbanPop",
TABLE = "Bzone",
GROUP = "Year",
TYPE = "people",
UNITS = "PRSN",
PROHIBIT = c("NA", "<= 0"),
ISELEMENTOF = ""
)
),The Set component contains one or more items describing datasets to be saved in the datastore. Each item in the Set component specifies attributes for one or more related datasets as follows:
- NAME: the names of one or more datasets to be saved;
- TABLE: the name of the table that the datasets are to be saved in;
- GROUP: the type of group where the table is located in the datastore (i.e. Global, Year, BaseYear);
- TYPE: the data type (e.g. double, distance, compound);
- UNITS: the measurement units for the data;
- NAVALUE: the value used to represent NA in the datastore;
- PROHIBIT: data values that are prohibited or “” if not applicable (see Section 8.1.1.1 for explanation);
- ISELEMENTOF: allowed categorical data values or “” if not applicable (see Section 8.1.1.1 for explanation);
- SIZE: the maximum number of characters for character data (or 0 for numeric data); and,
- DESCRIPTION: descriptions of the data corresponding to the names in the NAME attribute
The requirements for the UNITS attribute are the same as described above for Get component items. Following is an example of the Set component of the AssignRoadMilesSpecifications in the VETransportSupply package.
Set = items(
item(
NAME = "FwyLaneMiPC",
TABLE = "Marea",
GROUP = "Year",
TYPE = "compound",
UNITS = "MI/PRSN",
NAVALUE = -1,
PROHIBIT = c("NA", "< 0"),
ISELEMENTOF = "",
SIZE = 0,
DESCRIPTION = "Ratio of urbanized area freeway and expressway lane-miles to urbanized area population"
)
)It should be noted that it may not be possible to prespecify the SIZE attribute for a dataset. For example, if a unique household ID is assigned, the SIZE attribute will depend on the number of households and so must be calculated when the module is run. In such a circumstance, the SIZE attribute is omitted from the item and the module function must calculate it and include the calculated value as an attribute of the output dataset. The following code snippet from the CreateHouseholds module code in the VESimHouseholds package shows how this is done. In this example, the list of data that the module function returns is named Out_ls. The household ID (HhId) dataset is in the Household table of the Year group. The R attributes function is called to set the SIZE attribute of the HhId dataset equal to the maximum of the number of characters in each of the entries in the dataset.
attributes(Out_ls$Year$Household$HhId)$SIZE <- max(nchar(Out_ls$Year$Household$HhId)) The last component in the the module specifications list is the Call component. This is an optional component of the module specifications and is included if the module calls any other modules or if the module may be called by other modules. Following are Call component examples. The first is an example for a module that may be called. The second is an example for a module which calls another module. Section 8.1.2.2 explains module calling in detail.
#Call component for a module that may be called
Call = TRUE
#Call component for a module that calls another module
Call = items(
CalcDvmt = "VEHouseholdTravel::CalculateHouseholdDvmt"
)10.5.0.1.2.1 8.1.2.1 OPTIONAL Attribute
Module developers can use the OPTIONAL attribute to identify optional inputs or data to be retrieved from the datastore. This enables modules to be written to respond to optional inputs. For example, in the GreenSTEP and RSPM models, users may provide inputs on the average carbon intensity of fuels (grams CO2e per megajoule) by model run year. This allows users to model a scenario where state regulations require the average carbon intensity of fuels to be reduced over time. If the user supplies those data, the models calculate carbon emissions using those inputs. If not, the model calculates emissions using data on the carbon intensities of different fuel types and the mix of those fuel types.
If the OPTIONAL attribute for an item is missing, then the item is not optional. If the OPTIONAL attribute is not missing but is set equal to FALSE then the item is not optional as well. Only when the OPTIONAL attribute is present and set equal to TRUE does the framework regard the item as optional (OPTIONAL = TRUE).
If an input (Inp) item is identified as optional, the framework checks whether the identified input file is present. If the file is present, then the framework will process the data and load it into the datastore. Otherwise the item is ignored. Note that optional inputs can’t be combined with non-optional inputs in the same file. This will cause an error. Also not that since the framework does not automatically supply inputs to the module, there must be optional Get items corresponding to the optional Inp items. When the framework sees an optional Get item, it checks the datastore to see whether the optional dataset(s) are present. If so, it retrieves them.
10.5.0.1.2.2 8.1.2.2 CALL Specification
If same calculation code needs to be executed a number of times, it is best to define a function to encapsulate the code and then call the function whenever the calculation needs to be carried out. This reduces errors and code maintenance hassles. Likewise, module code duplication is reduced in the VisionEval model system by allowing modules to call other modules for their calculation services. For example, the BudgetHouseholdDvmt module in the VETravelPerformance package calls the CalculateAltModeTrips module in the VEHouseholdTravel package to recalculate trips by alternate modes (walk, bike, transit) to reflect budget-adjusted household DVMT. Calling a module is more involved than just calling the function that carries out the module’s calculations because the module function will not work unless it is supplied with the datasets identified in its Get specifications. Of course the calling module could include those specifications in its own specifications, but that would create a significant potential for coding errors and maintenance problems (e.g. if the called module is module is modified at a later time). For this reason the software design includes functionality for calling modules in a simple manner which leaves the data management details to the framework behind the scenes.
A module’s call status is specified in the Call component of the module specifications. There are 3 possibilities for a module’s call status. First, a module may be called by other modules. In this case the specification is Call = TRUE. Second, the module may call other modules. In this case the call specification is a list that identifies each of the modules that are called, assigning a reference to the module to an alias (i.e. alias = module) as shown in the following example:
Call = items(
CalcDvmt = "CalculateHouseholdDvmt",
ReduceDvmt = "ApplyDvmtReductions",
CalcVehTrips = "CalculateVehicleTrips",
CalcAltTrips = "CalculateAltModeTrips"
)The alias is the name that the called module will be referred to by the calling module code. The module is the name of the called module. The VisionEval framework software identifies which package the module resides in from the ModulesByPackage_df table in the ModelState_ls list. It is also possible to hard code the package name in the call definition. For example the CalcDvmt alias could be assign to VEHouseholdTravel::CalculateHouseholdDvmt. This is discouraged, however, because doing so limits the ability to maintain different versions of packages that have module modifications.
Call = items(
CalcDvmt = "VEHouseholdTravel::CalculateHouseholdDvmt"
)The third possibility is that a module may not be called and calls no other modules. In that case a Call component is not included in the module specifications.
There are some important restrictions to module calling. First, a module that may be called cannot call another module. If this restriction did not exist, there could be deeply nested module calls which could make debugging and understanding how a model works very difficult. Second, a module that may be called cannot have any inputs (i.e. Inp component). The reason for this restriction is that the function of called modules is to provide calculation services and those are hidden from model users. Unless a called module was also called directly in the model run script, there would be no way model user would know to supply input files without diving into the details about the module doing the calling.
The framework does the following when a module is run which calls one or more other modules. For each of the modules that are called, the framework:
1) Reads the ‘Get’ specifications for the called module, gets the datasets from the datastore, puts them in the standard list structure, and adds to the list that will be returned to the calling module as a component whose name is the assigned alias. In the example above, the retrieved datasets would be in a component named “CalcDvmt.”
2) Creates a list which holds the values of the called module functions. Each called module function is a component of the list whose name is the assigned alias. In the example above, this function list would have one component named “CalcDvmt” which contains the value of the CalculateHouseholdDVMT function.
3) The framework passes these two lists to the calling module when it is run. Thus the calling module function must be written to accept two arguments rather than one.
The calling module code invokes the called module by calling it from the function list that the framework passes it and passing to it the data it needs from the data list. Following from the example above, if the data list is called ‘L’ and the function list is called ‘M,’ then the CalculateHouseholdDVMT module function would be called in the AssignHhVehiclePowertrain module as follows:
M$CalcDvmt(L$CalcDvmt)10.5.0.1.3 8.1.3 Module Function
The function definitions section of the module script is used to define all functions that will be used to implement the module. One of these functions is the main function that is called by the software framework to run the module. This function must have the same name as the module name. For example, the main function of the CreateHouseholds module is named CreateHouseholds as well. This function must be written to accept one argument, a list, which by convention is named L if the module calls no other modules, and two arguments (as explained in the previous section) if the module calls other modules. This list contains all of the datasets identified in the Get component of the module data specifications. The structure of this list is described in Section 7. The main function returns a list which contains all of the datasets identified in the Set component of the module data specifications and structured as described in Section 7. The software framework includes a function, initDataList, to initialize an outputs list having the proper structure with Global, Year, and BaseYear components. The module code will fill in each component with components for each of the specified tables and datasets within those tables. Following is a simple example from the AssignTransitService module in the VETransportSupply package. First, here are the module’s Set items to provide context for the module code:
Set = items(
item(
NAME = "TranRevMiPC",
TABLE = "Marea",
GROUP = "Year",
TYPE = "compound",
UNITS = "MI/PRSN",
NAVALUE = -1,
PROHIBIT = c("NA", "< 0"),
ISELEMENTOF = "",
SIZE = 0,
DESCRIPTION = "Ratio of bus-equivalent revenue-miles (i.e. revenue-miles at the same productivity - passenger miles per revenue mile - as standard bus) to urbanized area population"
)
)The AssignTransitService function returns one dataset called TranRevMiPc. This dataset will be placed in the Marea table in the Year group (i.e. in the group for each model run year). Following is the function definition:
AssignTransitService <- function(L) {
#Set up
#------
#Fix seed as synthesis involves sampling
set.seed(L$G$Seed)
#Define vector of modes
Md <- as.character(BusEquivalents_df$Mode)
#Define vector of Mareas
Ma <- L$Year$Marea$Marea
#Calculate bus equivalent revenue miles
#--------------------------------------
#Make table of revenue miles by Marea
RevMi_df <- data.frame(L$Year$Marea[paste0(Md, "RevMi")])
colnames(RevMi_df) <- Md
rownames(RevMi_df) <- Ma
RevMi_MaMd <- as.matrix(RevMi_df)
#Calculate the bus equivalent revenue miles
BusEq_Md <- BusEquivalents_df$BusEquivalents
names(BusEq_Md) <- Md
BusEqRevMi_Ma <-
rowSums(sweep(RevMi_MaMd, 2, BusEq_Md, "*"))[Ma]
#Calculate the bus equivalent revenue miles per capita
#-----------------------------------------------------
#Calculate population in the urbanized area
UrbanPop_Ma <-
tapply(L$Year$Bzone$UrbanPop, L$Year$Bzone$Marea, sum)[Ma]
#Calculate Marea bus equivalent revenue miles per capita
TranRevMiPC_Ma <- BusEqRevMi_Ma / UrbanPop_Ma
#Return the results
#------------------
#Initialize output list
Out_ls <- initDataList()
Out_ls$Year$Marea <-
list(TranRevMiPC = TranRevMiPC_Ma)
#Return the outputs list
Out_ls
}A module should include procedures as necessary to check for errors in output datasets. In most cases this won’t be necessary if the module passes tests during development because the framework checks that all datasets passed to the module meet the module’s specifications. However there may be conditions where some combinations of acceptable inputs produce unacceptable outputs. In such cases, the module code should check for unacceptable outputs and if found, report those to the framework for reporting to the model user and stopping the model run. The module should not stop the model run. Instead it composes a descriptive error message and adds it to an Errors component of the module outputs list. If there is more than one error, the Errors component will be a vector of error messages. The framework writes the error message(s) to the model run log and stops model execution.
Warnings are handled in a similar way. If the module code is checks for warning conditions (i.e. model execution should not be stopped but users should be about the condition) and finds any, it composes a warning message that is added to a Warnings component of the module outputs list. The framework writes warnings messages to the model run log.
The functionality for processing module errors and warnings is used primarily by Initialize modules as explained in Section 8.1.5.
10.5.0.1.4 8.1.4 Module Documentation
It is recommended that module documentation be included at the top of the module script file if possible to make it easier to review and analyze the module. A standardized approach has been developed to do this that produces documentation in markdown format that can be viewed as a web page or converted to other formats for printing. At the end of the module script the documentModule function like this documentModule("PredictIncome"). The documentModule parses the module script, extracts the documentation block at the head of the file and inserts any text, tables, or figures that are saved by the script and inserted into the documentation using special tags as explained below. In addition, the documentModule function reads the module specifications and creates formatted tables showing module inputs, datasets used, and datasets produced. The documentation file(s) are then saved to a model_docs directory in the inst/extdata directory which is in turn a directory in the installed package. The documentModule function is called at the end of the script because the rest of the script must be executed to produce the datasets that are inserted into the documentation file. Following is an example of what a documentation block looks like.
#<doc>
## PredictIncome Module
#### September 6, 2018
#
#This module predicts the income for each simulated household given the number of workers in each age group and the average per capita income for the Azone where the household resides.
#
### Model Parameter Estimation
#Household income models are estimated for *regular* households and for *group quarters* households.
#
#The household income models are estimated using Census public use microsample (PUMS) data that are compiled into a R dataset (HhData_df) by the 'CreateEstimationDatasets.R' script when the VESimHouseholds package is built. The data that are supplied with the VESimHouseholds package downloaded from the VisionEval repository may be used, but it is preferrable to use data for the region being modeled. How this is done is explained in the documentation for the *CreateEstimationDatasets.R* script.
#
#The household income models are linear regression models in which the dependent variable is a power transformation of income. Power transformation is needed in order to normalize the income data distribution which has a long right-hand tail. The power transform is found which minimizes the skewness of the income distribution. The power transform for *regular* households is:
#
#<txt:HHIncModel_ls$Pow>
#
#The power transform for *group quarters* households is:
#
#<txt:GQIncModel_ls$Pow>
#
#The independent variables for the linear models are power transformed per capita income for the area, the number of workers in each of 4 worker age groups (15-19, 20-29, 30-54, 55-64), and the number of persons in the 65+ age group. In addition, power-transformed per capita income is interacted with each of the 4 worker groups and 65+ age group variable. The summary statistics for the *regular* household model are as follows:
#
#<txt:HHIncModel_ls$Summary>
#
#The summary statistics for the *group quarters* household model are as follows:
#
#<txt:GQIncModel_ls$Summary>
#
#An additional step must be carried out in order to predict household income. Because the linear model does not account for all of the observed variance, and because income is power distribution, the average of the predicted per capita income is less than the average per capita income of the population. To compensate, random variation needs to be added to each household prediction of power-transformed income by randomly selecting from a normal distribution that is centered on the value predicted by the linear model and has a standard deviation that is calculated so as the resulting average per capita income of households match the input value. A binary search process is used to find the suitable standard deviation. Following is the comparison of mean values for the observed *regular* household income for the estimation dataset and the corresponding predicted values for the estimation dataset.
#
#<tab:HHIncModel_ls$MeanCompare>
#
#The following figure compares the distributions of the observed and predicted incomes of *regular* households.
#
#<fig:reg-hh-inc_obs-vs-est_distributions.png>
#
#Following is the comparison of mean values for the observed *group quarters* household income for the estimation dataset and the corresponding predicted values for the estimation dataset.
#
#<tab:GQIncModel_ls$MeanCompare>
#
#The following figure compares the distributions of the observed and predicted incomes of *groups quarters* households.
#
#<fig:gq-hh-inc_obs-vs-est_distributions.png>
#
### How the Module Works
#This module runs at the Azone level. Azone household average per capita income and group quarters average per capita income are user inputs to the model. The other model inputs are in the datastore, having been created by the CreateHouseholds and PredictWorkers modules. Household income is predicted separately for *regular* and *group quarters* households. Per capita income is transformed using the estimated power transform, the model dependent variables are calculated, and the linear model is applied. Random variation is applied so that the per capita mean income for the predicted household income matches the input value.
#
#</doc>
There are several things to note about this example. The first is that all text in the block is commented out (preceded by #). Since R doesn’t support block comments, each line must be commented. It may not look this way in the example because of word wrapping, but every line is commented. Second, the start and end of the documentation block are denoted by matching <doc> and </doc> tags. The parser uses these to extract the documentation from the script. After the document has been extracted, the leading comments are stripped off, resulting in markdown-formatted text. The other comment (#) symbols in the text are actually markdown formatting to identify headings of different levels. Documentation can include any standard markdown formatting such as emphasis, links, and tables. In addition, the documentation can include special tags as shown in the example. Three types of tags are available:
<txt:xxxx>inserts a block of text that is contained in the referenced object. For example the<txt:GQIncModel_ls$Summary>tag in the example will insert summary statistics for the group quarters income model.<tab:xxxx>inserts data that can be presented in table for such as a data frame. For example the<tab:HHIncModel_ls$MeanCompare>tag in the example will insert a table which compares observed and estimated mean values.<fig:xxxx>creates a markdown reference to an image file so that it will be show in the proper place when the markdown is displayed in a browser or converted to another document form. For example the<fig:reg-hh-inc_obs-vs-est_distributions.png>tag in the example will insert a figure which compares observed and estimated income distributions when the markdown is displayed.
It is helpful to include test code in the module script to aid with module development. The framework includes a testModule function to assist with module testing. This function is described in detail in Section 9.2.1. Testing requires having sample input files containing datasets specified by the modules Inp specifications. These are stored in the ‘inputs’ directory of the ‘tests’ directory. The ‘tests’ directory also must contain a ‘defs’ directory which contains all the required model definitions files (see Section 6.1). Finally, the ‘tests’ directory must contain a datastore of the type specified in the ‘run_parameters.json’ file, which contains all of the datasets specified in the modules Get specifications. In the first stage of module testing, the module specifications are checked, all input files are checked, the presence of all required data is checked, and an input list (L) is returned for use in module development. For this test, the DoRun argument of the testModule function needs to be set to FALSE. After the module code has been written, the module is tested again to check whether the module code is working correctly and that it returns results that are consistent with the Set specifications for the module. For this test, the DoRun argument of the testModule function must be TRUE. After the module has been tested, it is important to comment out all the testing code in the script because it must not be run when the package is built. Following is an example of testing code in a module that has been commented out.
10.5.0.1.5 8.1.5 Initialize Module
Although the framework performs several checks on module input data based on the module specifications, there will be times when additional checks of inputs will be necessary and possibly transformations as well. For example, several datasets could have proportions data that must add up to 1 across the datasets. For example, 4 input datasets for the PredictHousing module of the *VELandUse** package give the proportions of households in each Bzone that are in each of 4 income quartiles for the Azone that the Bzones are located in. These inputs should be checked to assure that the sum of all quartile proportions for each Bzone adds up to one. If any sums are not close to 1, then the model user needs to be alerted to the fact so that they can correct the input file. If all the sums are close to 1 but some are not exactly 1 (due to rounding errors in preparing inputs), the inputs should be automatically adjusted to equal 1 before they are saved in the datastore. If a module developer needs to establish more complex checks and transformations like this, they do so in a special module that they name Initialize.
In the Initialize module, the module specifications identify all of the input datasets that need to be checked. This is done in same manner as described in Section 8.1.2. The module specifications will not have Get or Set components since the only purpose of the Initialize module is to process inputs. There can only be one Initialize module in a package and so all inputs that need additional checking, regardless of which module in the package will use them, need to be processed in the Initialize module. Datasets that are listed in the Inp specifications of the Initialize module must not be included in the Inp specifications of any other module in the package.
When a model is intialized by the initializeModel function in the run_model.R script (Section 5 and Appendix A), each of the module packages that will be run by the script is checked for the presence of an Initialize module. Any that are found are added to a list of modules that require input processing. When the initializeModel function processes the inputs for an Initialize module it does so in two steps. In the first step, it performs the standard input processing that is done for all modules (input files are read and datasets are checked for completeness and correctness). The output of this checking process is a standard outputs list with Global, BaseYear, and Year components. In the second step, the framework calls the Initialize module function and passes it the outputs list from the first step. The Initialize module does whatever enhanced data checking and transformation is necessary and returns an outputs list having the same structure as the inputs list with the addition of Errors and Warnings components (see Section 8.1.3). Several of the VisionEval packages include Initialize modules and can be used as examples.
10.5.0.2 8.2. The inst/extdata Directory
By convention, the ‘inst/extdata’ directory is the standard place to put external (raw) data files as opposed to R datasets which are placed in the ‘data’ directory. This is where most model estimation data is kept. Section 8.1.1.1 provides a detailed explanation. The directory should include a subdirectory named ‘sources’ to hold reports or other external documentation if needed.
10.5.0.3 8.3. The data-raw Directory
In some instances it is impractical to include the model estimation data as files in the ‘inst/extdata’ directory, and not necessary for model users to provide regional model estimation data. If, for example, the source data has confidential elements, it may be necessary to preprocess the data to anonymize it before including in the package; or the source data may be too large to include as a text file in the package. In these cases, source data may be processed outside of the package and then the processed datasets included in the package as datasets in R binary files. If that is done, the binary data files should be placed in a directory named ‘data-raw’ in the package.
10.5.0.4 8.4. The tests Directory
The ‘tests’ directory contains R scripts and the results of module tests. The scripts directory contains all the scripts used to carry out module tests. The directory also contains subdirectories for each of the model types the module is to be tested with (e.g VE-RSPM, VE-State, VE-RPAT). Two approaches are available for handing module data that includes input files the module uses, all the defs files, and a datastore which contains all the datasets used by the module aside from those in the input files. These data can be included in the package or they may be kept in a central repository. If they are included in the package, they must be placed in the directory for the corresponding model. This is necessary to avoid conflicts in the test data for different models. The scripts directory includes a testing script which runs the tests on all modules in a package for a particular module. For example, the script for testing modules in a VE-State application is named vestate_test.R. The scripts directory also includes a test.R script which calls the individual model test scripts for automated package testing. If the centralized data testing approach is used, a test_functions.R script needs to be included. This includes functions needed to support the centralized test data approach. The test process is still not finalized. In the future this functionality will be included in the framework software. Following is an example of a test script using the central data approach:
#vestate_test.R
#--------------
#Load packages and test functions
library(visioneval)
library(filesstrings)
source("tests/scripts/test_functions.R")
#Define test setup parameters
TestSetup_ls <- list(
TestDataRepo = "../Test_Data/VE-State",
DatastoreName = "Datastore.tar",
LoadDatastore = FALSE,
TestDocsDir = "vestate",
ClearLogs = TRUE,
# SaveDatastore = TRUE
SaveDatastore = FALSE
)
#Define the module tests
Tests_ls <- list(
list(ModuleName = "CreateHouseholds", LoadDatastore = FALSE, SaveDatastore = TRUE, DoRun = TRUE),
list(ModuleName = "PredictWorkers", LoadDatastore = TRUE, SaveDatastore = TRUE, DoRun = TRUE),
list(ModuleName = "AssignLifeCycle", LoadDatastore = TRUE, SaveDatastore = TRUE, DoRun = TRUE),
list(ModuleName = "PredictIncome", LoadDatastore = TRUE, SaveDatastore = TRUE, DoRun = TRUE)
)
#Set up, run tests, and save test results
setUpTests(TestSetup_ls)
doTests(Tests_ls, TestSetup_ls)
saveTestResults(TestSetup_ls)
Section 9.2.1 provides more information on using the testModule function.
10.6 Software Framework
The software framework for the VisionEval model system is implemented by a set of functions contained in the visioneval package. These functions are classified in four groups: user, developer, control, and datastore. Model user functions are those used to write scripts to run VisionEval models. Section 9.1 describes how these are used. Appendix G contains full documentation of model user functions. The package contains contains standard documentation for all of the functions. Module developer functions are those that module developers may call in their module code or that otherwise aid in module developing and testing. Section 9.2 describes the most important module development functions and Appendix H includes full documentation of all the functions. Most of the functions in the VisionEval framework are functions that control the initialization of a VisionEval model run and the execution of VisionEval modules. These functions are internal to the VisionEval software framework and are not to be used by model users or module developers. Section 9.3 provides an overview of these functions and Appendix I includes full function documentation. The remaining functions are functions which directly interact with the model datastore. Section 9.4 provides an overview of these functions and Appendix J provides full documentation.
Additional documentation which shows the calling relationships between functions is [no longer, 9/17/2019] available in a interactive visualization. This visualization shows the names of the functions as nodes in a network graph with arrows connecting the nodes showing the functions that are called by each function (arrows point from the calling function to the called function). The nodes are colored-coded to indicate the function groups: blue indicates model user functions, green indicates module developer functions, yellow indicates framework control functions, and red indicates datastore interaction functions. Clicking on a function in the visualization highlights the function and all the arrows connected to it. It also provides summary information about the function including a description of what it does, descriptions of all the function arguments, and a description of the function’s return value.
10.6.0.1 9.1. API for Model Users
Three functions are part of the API for model users: ‘initializeModel,’ ‘runModule,’ and ‘getYears.’ These are explained below in turn.
The ‘initializeModel’ function prepares the model for running modules. This includes:
1) Creating the “ModelState.Rda” file that contains global parameters for the model run and variables used to keep track of the state of the datastore and other aspects of the model run (Section 6.6);
2) Creating a log file that is used to record model status messages such as warning and error messages;
3) Creating and initializing the model datastore including loading a specified datastore;
4) Processing the model geography definition file and setting up the appropriate geographic tables in the datastore;
5) Checking whether all the specified module packages are installed and that all module specifications are correct;
6) Parsing the “run_model.R” script and simulating the model run to confirm that the datastore will contain the data that each module needs when it is called and that the data specifications are consistent with the module ‘Get’ specifications;
7) Checking whether all the scenario input files identified by the specified modules are present, and that the data are consistent with specifications; and,
8) Loading all the data in the input files into the datastore.
If any errors are found during the model initialization process, an error message will be displayed in the console and the initialization process will terminate. Detailed error messages in the log will identify the specific causes of errors. If the initialization proceeds without errors, the user can be assured that the model will run without errors. Following is a typical ‘initializeModel’ function call in a model run script.
initializeModel(
ParamDir = "defs",
RunParamFile = "run_parameters.json",
GeoFile = "geo.csv",
ModelParamFile = "model_parameters.json",
LoadDatastore = FALSE,
DatastoreName = NULL,
SaveDatastore = TRUE
)The function arguments and their meanings are as follows:
- ParamDir A string identifying the relative or absolute path to the directory where the parameter and geography definition files are located. The default value is “defs.”
- RunParamFile A string identifying the name of a JSON-formatted text file that contains parameters needed to identify and manage the model run. The default value is “run_parameters.json.”
- GeoFile A string identifying the name of a text file in comma-separated values format that contains the geographic specifications for the model. The default value is “geo.csv.”
- ModelParamFile A string identifying the name of a JSON-formatted text file that contains global model parameters that are important to a model and may be shared by several modules.
- LoadDatastore A logical identifying whether an existing datastore should be loaded.
- DatastoreName A string identifying the full path name of a datastore to load or NULL if an existing datastore in the working directory is to be loaded.
- SaveDatastore A string identifying whether if an existing datastore in the working directory should be saved rather than removed.
Some consistency checks are done if a datastore is specified to be loaded including: - The base year for the loaded datastore must be the same as defined for the model run. - The defined geography for the loaded datastore must be the same as defined for the model run.
As the name suggests, the ‘runModule’ function runs a module. Following is an example of how it is invoked:
runModule(ModuleName = "CreateHouseholds",
PackageName = "VESimHouseholds",
RunFor = "AllYears",
RunYear = Year)The function arguments and their meanings are as follows:
- ModuleName A string identifying the name of a module object.
- PackageName A string identifying the name of the package the module is a part of.
- RunFor A string identifying whether to run the module for all years (AllYears), only the base year (BaseYear), or for all years except the base year (NotBaseYear).
- RunYear A string identifying the run year.
The runModule function runs the named module within the runModule function environment. This is a significant improvement over how functions that implement the submodels in the current GreenSTEP (RSPM, EERPAT, RPAT) are run. In these models, functions are run in the the global environment. As a consequence, the global environment collects objects that increase the potential for name conflicts if care is not taken to keep it clean. By running modules within the runModule function environment, no changes are made to the global environment and all objects that are created in the process vanish when the runModule function completes the work of running the module.
Modules can be run for multiple years by running them in a loop which iterates through all of the years identified in the ‘Years’ parameter specified in the “run_parameters.json” file (Section 6.1). Section 5 shows an example of using such a loop. Rather than hard code the model run years in the loop, the user can use the ‘getYears’ function to query and return the vector of years.
10.6.0.2 9.2. API for Module Developers
The VisionEval API for module developers currently includes 10 functions. These are presented below in 3 groups:
- Key module script functions;
- Functions to help developers write specifications that are consistent with other modules; and,
- Functions that developers may use to simplify model implementation.
10.6.0.2.1 9.2.1. Key Module Script Functions
Four functions are almost always used in module scripts.
Module specifications are written as nested R lists that are structured in a particular way (Section 8). Rather than use the ‘list’ function to define the list structure, two alias functions - ‘item’ and ‘items’ - are used to define the structure. An example of how these functions are used is shown in Appendix E. Although modules will run if the ‘list’ function is used instead, it is highly recommended that ‘item’ and ‘items’ be used to maintain a consistent style for all modules.
The ‘processEstimationInputs’ function must be used if a module includes procedures for estimating model parameter(s) from regional data (Section 8). This function is used to check that the data supplied to calculate the regional parameter(s) are consistent with specifications. The function arguments are as follows:
- Inp_ls A list that describes the specifications for the estimation file. This list must meet the framework standards for specification description.
- FileName A string identifying the file name. This is the file name without any path information. The file must located in the “inst/extdata” directory of the package.
- ModuleName A string identifying the name of the module the estimation data is being used in.
The function returns a data frame containing the estimation inputs if all the supplied specifications are met. If any of the specifications are not met, an error is thrown and details regarding the specification error(s) are written to the console.
The ‘testModule’ function is an essential tool for testing that the module will work correctly in the VisionEval model system. The test module function tests a module with a test setup that mimics a model run. A test datastore needs to be present unless no data from other modules is needed (i.e. all data used by the module is supplied by input data). All inputs required by the module must be present, and all the standard model definitions files included in the “defs” directory (Section 6.1) must be present as well. When this function is invoked, the following tests are done on a module:
- Checks whether module specifications are proper;
- Checks whether test module inputs are consistent with the module ‘Inp’ specifications and that they can be loaded into a test datastore;
- Checks whether the test datastore with the loaded inputs contains all the data needed for the module to run;
- Checks whether the module will run without error; and,
- Checks whether the outputs of the module are consistent with the module ‘Set’ specifications.
The function arguments are as follows:
- ModuleName A string identifying the module name.
- ParamDir A string identifying the location of the directory where the run parameters, model parameters, and geography definition files are located. The default value is defs. This directory should be located in the tests directory.
- RunParamFile A string identifying the name of the run parameters file. The default value is run_parameters.json.
- GeoFile A string identifying the name of the file which contains geography definitions.
- ModelParamFile A string identifying the name of the file which contains model parameters. The default value is model_parameters.json.
- LoadDatastore A logical value identifying whether to load an existing datastore. If TRUE, it loads the datastore whose name is identified in the run_parameters.json file. If FALSE it initializes a new datastore.
- SaveDatastore A logical value identifying whether the module outputs will be written to the datastore. If TRUE the module outputs are written to the datastore. If FALSE the outputs are not written to the datastore.
- DoRun A logical value identifying whether the module should be run. If FALSE, the function will initialize a datastore, check specifications, and load inputs but will not run the module. It will return the list of module inputs. This is described in more detail below. If TRUE, the module will be run and results will be checked for consistency with the module’s ‘Set’ specifications.
- RunFor A string identifying whether to run the module for all years (AllYears), only the base year (BaseYear), or for all years except the base year (NotBaseYear)
- StopOnErr A logical value indicating whether model execution should be stopped if the module transmits one or more error messages or whether execution should continue with the next module. The default value is TRUE. This is how error handling will ordinarily proceed during a model run. A value of FALSE is used when ‘Initialize’ modules in packages are run during model initialization. These ‘Initialize’ modules are used to check and preprocess inputs. For this purpose, the module will identify any errors in the input data, the ‘initializeModel’ function will collate all the data errors and print them to the log.
- RequiredPackages A string vector identifying any other VisionEval packages that the module calls modules from or access datasets from.
- TestGeoName A string identifying the name of a geographic area to return the data for. If the DoRun argument is FALSE, the function returns a list containing all the data the module requests. It will only return the data for one geographic area in the set identified in the RunBy specification. For example if the RunBy specification is Marea, the function will return the list of data for only one Marea. This argument can be used to specify which geographic area the data is to be returned for. Otherwise the data for the first area identified in the datastore is returned.
If the DoRun argument is TRUE, the module will be run and there will be no return value. The module will run for all geographic areas and the outputs will be checked for consistency with the module’s Set specifications. If that argument is FALSE, the return value of the function is a list containing the data identified in the module’s Get specifications. That setting is useful for module development in order to create the all the data needed to assist with module programming. It is used in conjunction with the ‘getFromDatastore’ function to create the dataset that will be provided by the framework. The example module script in Appendix E shows how this aspect of the ‘testModule’ function can be used by module developers to make the development of their code easier. The function also writes out messages to the console and to the log as the testing proceeds. These messages include the time when each test starts and when it ends. When a key test fails, requiring a fix before other tests can be run, execution stops and an error message is written to the console. Detailed error messages are also written to the log.
10.6.0.2.2 9.2.2. Functions to Assist Specification Writing
As was explained above in Sections 4.1 and 8.1, the VisionEval model system uses data specifications to help assure that modules can work properly with one another. The data specifications are saved as attributes for each dataset that are saved to the datastore by a module. The specifications are checked for consistency for each dataset a module requests to be retrieved from the datastore. A couple of functions assist a module developer with identifying datasets that registered modules produce and for retrieving ‘Get’ specifications for the datasets the developer’s module will use.
The ‘item’ and ‘items’ functions are used to organize specifications in the module script. They are aliases of the R language ‘list’ function.
The ‘readVENameRegistry’ function returns a list containing the specifications for all datasets that registered modules save to the datastore. This list contains two components. The components are data frames containing the specifications for all datasets identified in the ‘Inp’ and ‘Set’ of registered modules. Each data frame row lists the specifications for a dataset as well as the module which produces the dataset and the package the module is in. This function is useful to developers for:
- Avoiding dataset naming conflicts with other modules; and,
- Identifying datasets produced by other modules that can be used in module calculations.
At the present time, the ‘readVENameRegistry’ function has fairly rudimentary functionality. The only argument, ‘NameRegistryDir,’ allows the user to specify the local directory where the name registry is located. In the future, the function will read the registry from the remote repository where VisionEval modules are stored. Also, the function will also be modified to enable the module developer to search for datasets based on keywords, module names, package names, and keywords.
The ‘getRegisteredGetSpecs’ function helps the module developer to write ‘Get’ specifications that are consistent with the specifications of registered datasets. This function returns a data frame containing the ‘Get’ specifications for specified datasets. The function arguments are as follows:
- Names_ A character vector of the dataset names to get specifications for.
- Tables_ A character vector of the tables that the datasets are a part of.
- Groups_ A character vector of the groups that the tables are a part of.
- NameRegistryDir A string identifying the path to the directory where the name registry file is located.
At the present time, the function returns a data frame which contains the ‘Get’ specifications for each requested dataset. It is up to the module developer to put the information into the proper form in the module script. In the future, the function will be modified to return the ‘Get’ specifications in list form that may be copied into a module script.
10.6.0.2.3 9.2.3. Processing Runtime Errors and Warnings
Section 8.1.5 described how the Initialize module can be used to perform input data checks more than the basic data checking performed by the framework. Although the use of framework data checks and an Initialize module can and should ideally catch all data errors/warnings during model initialization, there are sometimes instances where errors/warnings can only be caught during runtime. Runtime errors/warnings should ideally be caught gracefully and the user should be presented with an understandable message that will direct them to a solution rather than cause the program to crash with a cryptic R language message. The framework provides functions to assist the module developer to identify errors and warnings to the framework during runtime and to include helpful messages. The addErrorMsg function is used to pass an error message to the framework. If the module code catches a specified error, it should handle the error so that the code doesn’t crash and use the addErrorMsg function to record the error for processing by the framework. By default, if an error message is returned by a module, the runModule function will stop model execution, will print a message to the console that an error occurred in running the module, and will print details of the error that the module developer included in the message. The addErrorMsg function takes two arguments: ResultsListName, a string identifying the name of the results list returned by the module (e.g. ‘Out_ls’); ErrMsg a string that is the error message that will be written to the log for the user to see. Likewise, the addWarningMsg function likewise is used to pass a warning message to the framework. In the case of warnings, the model run is not stopped but a warning is printed in the console and details are written to the log. Following is an example of the use of addWarningMsg message in checking for unlikely population density:
#Calculate density measures
#--------------------------
#Population density
D1B_ <- with(D_df, Pop / Area)
#Check for high population density values and add warning
IsHighDensity_ <- D1B_ > 100
HighDensityBzones_ <- Bz[IsHighDensity_]
if (any(IsHighDensity_)) {
Msg <- paste0(
"The following Bzones in the year ", L$G$Year, " ",
"have population densities greater than ",
"100 persons per acre: ", paste(HighDensityBzones_, collapse = ", "), ". ",
"This density is a relatively high level. ",
"Check your Bzone area and housing inputs for these Bzones and make ",
"sure that they are correct."
)
addWarningMsg("Out_ls", Msg)
rm(Msg)
}
rm(IsHighDensity_, HighDensityBzones_)
#Employment density
D1C_ <- with(D_df, TotEmp / Area)
#Activity density
D1D_ <- with(D_df, (TotEmp + NumHh) / Area)10.6.0.2.4 9.2.4. Utility Functions for Implementing Modules
Many submodels of the GreenSTEP and RSPM models are linear or binomial logit models. Several of the binary logit model implementations adjust the constant to match specified input proportions. For example, the light truck model enables model users to specify a future light truck proportion and the model will adjust the constant to match that proportion. Likewise, several linear models adjust a dispersion parameter to match a specified population mean. This is done for example in the household income model to match future per capita income projections. The adjustments are made with the use of a binary search algorithm. The following three functions simplify the implementation of those models in the VisionEval model system.
The ‘applyLinearModel’ function applies a linear model and optionally adjusts the model to match a target mean value. It has the following arguments:
- Model_ls A list which contains the following components: 1) Type - which has a value of ‘linear’; 2) Formula - a string representation of the model equation; 3) PrepFun - a function which prepares the input data frame for the model application. If no preparation, this element of the list should not be present or should be set equal to NULL; 4) SearchRange - a two-element numeric vector which specifies the acceptable search range to use when determining the dispersion factor. 5) OutFun a function that is applied to transform the results of applying the linear model. For example to untransform a power-transformed variable. If no transformation is necessary, this element of the list should not be present or should be set equal to NULL.
- Data_df A data frame containing the data required for applying the model.
- TargetMean A number identifying a target mean value to be achieved or NULL if there is no target.
- CheckTargetSearchRange A logical identifying whether the function is to only check whether the specified ‘SearchRange’ for the model will produce acceptable values (i.e. no NA or NaN values). If FALSE (the default), the function will run the model and will not check the target search range.
It is important to note that the ‘Model_ls’ argument is a list that must contain the components listed above. Also, the ‘CheckTargetSearchRange’ argument must NOT be set equal TRUE in the call in the module function. Setting it equal to TRUE is only useful during model estimation to help set the target search range values.
The function returns a vector of numeric values for each record of the input data frame if the model is being run, or if the function is run to only check the target search range, a summary of predicted values when the model is run with dispersion set at the high value of the search range.
The ‘applyBinomialModel’ function applies a binomial model and optionally adjusts the model to match a target proportion. It has the following arguments which are similar to those of the ‘applyLinearModel’ function:
- Model_ls A list which contains the following components: 1) Type - which has a value of ‘binomial’; 2) Formula - a string representation of the model equation; 3) Choices - a two-element vector listing the choice set. The first element is the choice that the binary logit model equation predicts the odds of; 4) PrepFun - a function which prepares the input data frame for the model application. If no preparation, this element of the list should not be present or should be set equal to NULL; 5) SearchRange - a two-element numeric vector which specifies the acceptable search range to use when determining the factor for adjusting the model constant.
- Data_df A data frame containing the data required for applying the model.
- TargetProp A number identifying a target proportion for the default choice to be achieved for the input data or NULL if there is no target proportion to be achieved.
- CheckTargetSearchRange A logical identifying whether the function is to only check whether the specified ‘SearchRange’ for the model will produce acceptable values (i.e. no NA or NaN values). If FALSE (the default), the function will run the model and will not check the target search range.
ApplyRandom A logical value which determines how the binomial choice is made. The binomial choice model returns a probability that modeled selection is chosen. For example a housing type model could predict the probability that a household lives in a single-family home. If the ApplyRandom argument is TRUE, the function takes a sample from a uniform distribution from 0 to 1 and if the value is less than the probability the modeled choice is selected. Otherwise the alternate choice is selected. If the ApplyRandom argument is FALSE the modeled choice is selected if the modeled probability is greater than 0.5.
ReturnProbs A logical value which if TRUE returns the modeled choice probabilities instead of the modeled choices.
The function returns a vector of choice values for each record of the input data frame if the neither the CheckTargetSearchRange or ReturnProbs arguments are TRUE. If the ReturnProbs argument is TRUE the choice probabilities are returned. If the CheckTargetSearchRange argument is TRUE the function is run to only check the target search range, a two-element vector identifying if the search range produces NA or NaN values.
The third function, ‘binarySearch,’ is called by the ‘applyLinearModel’ function if the value of the ‘TargetMean’ argument is not NULL, and called by the ‘applyBinomialModel’ function if the value of the ‘TargetProp’ argument is not NULL. Module developers may find this function to be useful in their own module implementation code. The arguments of the function are:
- Function A function which returns a value which is compared to the Target argument. The function must take as its first argument a value which from the SearchRange_. It must return a value that may be compared to the Target value.
- SearchRange_ A two element numeric vector which has the lowest and highest values of the parameter range within which the search will be carried out.
- … One or more optional arguments for the Function.
- Target A numeric value that is compared with the return value of the ‘Function.’
- MaxIter An integer specifying the maximum number of iterations for the search to attempt in order to match the ‘Target’ within the specified ‘Tolerance.’
- Tolerance A numeric value specifying the proportional difference between the ‘Target’ and the return value of the Function to determine when the search is complete.
The function returns a value within the ‘SearchRange_’ for the function parameter which matches the target value.
Developers can refer to the source code for the ‘applyLinearModel’ and ‘applyBinomialModel’ functions to help understand how to use this function.
10.6.0.2.5 9.2.5. Module Documentation Function
Section 8.1.4 describes how module documentation is to be included in the module script. The ‘documentModule’ prepares formatted documentation from the script documentation. Refer to that section for more details.
10.6.0.2.6 9.2.6. Debugging Runtime Errors
Although the large majority of runtime errors are avoided by the framework’s data specification system, Initialize modules, and internal module error checks, inevitably some errors are only caught during runtime. This can pose challenges for debugging because the R language error messages can be cryptic. Often the easiest way to find the error is to recreate the state of the model at the start of module execution and then step through the module code to find the error. This is made simpler by using the fetchModuleData function. This function retrieves from the datastore the data that was passed to the module so that the code can be checked. The function takes the following arguments: - ModuleName: a string identifying the name of the module. - PackageName: a string identifying the name of the package that the module is in. - Year: a string identifying the model run year to retrieve the data for. - Geo: a string identifying the geography to retrieve the data for if the module’s ‘RunBy’ specification is not ‘Region.’ This argument is omitted if the ‘RunBy’ specification is ‘Region.’
Here is an example of the steps that were used in debugging the CalculateUrbanMixMeasure module which errored when running a model in the R console:
1. The model run stopped and a cryptic R error message was printed to the console.
2. The following code was entered at the console to get the dataset used by the module: L <- fetchModuleData("CalculateUrbanMixMeasure", "VELandUse", "2040"). Note that L is the standard name used by modules to refer to the data list provided by the framework to the module.
3. The module function code was retrieved with edit(VELandUse::CalculateUrbanMixMeasure) and pasted into a temporary file.
4. The module code in the temporary file was stepped through line by line until an error occurred. Then the values of data used in the computation could be checked to determine the cause. In this case, error was caused by very high population densities that were outside the range of values used to estimate the binomial logit model which determines the odds that a Bzone is urban-mixed-use. Calculating the odds involves exponentiation which in the case of very high densities produced infinite values which in turn caused probability values of NA. That in turn caused an error when used in the sample function.
5. Code to fix the module was developed and tested in the temporary file. Then fixed code was used to modify the CalculateUrbanMixMeasure code in the VELandUse package.
10.6.1 Appendix A: Example Model Run Script
#===========
#run_model.R
#===========
#This script demonstrates the VisionEval framework for the RSPM model.
#Load libraries
#--------------
library(visioneval)
#Initialize model
#----------------
initializeModel(
ParamDir = "defs",
RunParamFile = "run_parameters.json",
GeoFile = "geo.csv",
ModelParamFile = "model_parameters.json",
LoadDatastore = FALSE,
DatastoreName = NULL,
SaveDatastore = TRUE
)
#Run all demo module for all years
#---------------------------------
for(Year in getYears()) {
runModule(ModuleName = "CreateHouseholds",
PackageName = "VESimHouseholds",
RunFor = "AllYears",
RunYear = Year)
runModule(ModuleName = "PredictWorkers",
PackageName = "VESimHouseholds",
RunFor = "AllYears",
RunYear = Year)
runModule(ModuleName = "AssignLifeCycle",
PackageName = "VESimHouseholds",
RunFor = "AllYears",
RunYear = Year)
runModule(ModuleName = "PredictIncome",
PackageName = "VESimHouseholds",
RunFor = "AllYears",
RunYear = Year)
runModule(ModuleName = "PredictHousing",
PackageName = "VESimHouseholds",
RunFor = "AllYears",
RunYear = Year)
runModule(ModuleName = "LocateHouseholds",
PackageName = "VELandUse",
RunFor = "AllYears",
RunYear = Year)
runModule(ModuleName = "LocateEmployment",
PackageName = "VELandUse",
RunFor = "AllYears",
RunYear = Year)
runModule(ModuleName = "AssignDevTypes",
PackageName = "VELandUse",
RunFor = "AllYears",
RunYear = Year)
runModule(ModuleName = "Calculate4DMeasures",
PackageName = "VELandUse",
RunFor = "AllYears",
RunYear = Year)
runModule(ModuleName = "CalculateUrbanMixMeasure",
PackageName = "VELandUse",
RunFor = "AllYears",
RunYear = Year)
runModule(ModuleName = "AssignTransitService",
PackageName = "VETransportSupply",
RunFor = "AllYears",
RunYear = Year)
runModule(ModuleName = "AssignRoadMiles",
PackageName = "VETransportSupply",
RunFor = "AllYears",
RunYear = Year)
runModule(ModuleName = "AssignVehicleOwnership",
PackageName = "VEVehicleOwnership",
RunFor = "AllYears",
RunYear = Year)
runModule(ModuleName = "CalculateHouseholdDVMT",
PackageName = "VETravelDemand",
RunFor = "AllYears",
RunYear = Year)
runModule(ModuleName = "CalculateAltModeTrips",
PackageName = "VETravelDemand",
RunFor = "AllYears",
RunYear = Year)
}10.6.2 Appendix B: Geography Specification File (geography.csv) Examples
Figure A1. Example of a geography.csv file that only specifies Azones
Figure A2. Example of geography.csv file that specifies Azones and Bzones
10.6.3 Appendix C: Recognized Data Types and Units
Recognized Data Types and Units are defined in the ‘Types’ function. The definition also includes the factors for converting between units. This function definition is listed below:
$double
$double$units
[1] NA
$double$mode
[1] "double"
$integer
$integer$units
[1] NA
$integer$mode
[1] "integer"
$character
$character$units
[1] NA
$character$mode
[1] "character"
$logical
$logical$units
[1] NA
$logical$mode
[1] "logical"
$compound
$compound$units
[1] NA
$compound$mode
[1] "double"
$currency
$currency$units
$currency$units$USD
USD
1
$currency$mode
[1] "double"
$distance
$distance$units
$distance$units$MI
MI FT KM M
1.00000 5280.00000 1.60934 1609.34000
$distance$units$FT
MI FT KM M
0.000189394 1.000000000 0.000304800 0.304800000
$distance$units$KM
MI FT KM M
0.621371 3280.840000 1.000000 1000.000000
$distance$units$M
MI FT KM M
0.000621371 3.280840000 0.001000000 1.000000000
$distance$mode
[1] "double"
$area
$area$units
$area$units$SQMI
SQMI ACRE SQFT SQM HA SQKM
1.00000e+00 6.40000e+02 2.78800e+07 2.59000e+06 2.58999e+02 2.58999e+00
$area$units$ACRE
SQMI ACRE SQFT SQM HA SQKM
1.56250e-03 1.00000e+00 4.35600e+04 4.04686e+03 4.04686e-01 4.04686e-03
$area$units$SQFT
SQMI ACRE SQFT SQM HA SQKM
3.5870e-08 2.2957e-05 1.0000e+00 9.2903e-02 9.2903e-06 9.2903e-08
$area$units$SQM
SQMI ACRE SQFT SQM HA SQKM
3.86100e-07 2.47105e-04 1.07639e+01 1.00000e+00 1.00000e-04 1.00000e-06
$area$units$HA
SQMI ACRE SQFT SQM HA SQKM
3.86102e-03 2.47105e+00 1.07639e+05 3.86102e-03 1.00000e+00 1.00000e-02
$area$units$SQKM
SQMI ACRE SQFT SQM HA SQKM
3.86102e-01 2.47105e+02 1.07600e+07 1.00000e+06 1.00000e+02 1.00000e+00
$area$mode
[1] "double"
$mass
$mass$units
$mass$units$LB
LB TON MT KG GM
1.00000e+00 5.00000e-04 4.53592e-04 4.53592e-01 4.53592e+02
$mass$units$TON
LB TON MT KG GM
2.00000e+03 1.00000e+00 9.07185e-01 9.07185e+02 9.07185e+05
$mass$units$MT
LB TON MT KG M
2.20462e+03 1.10231e+00 1.00000e+00 1.00000e+03 1.00000e+06
$mass$units$KG
LB TON MT KG GM
2.20462e+00 1.10231e-03 1.00000e-03 1.00000e+00 1.00000e+03
$mass$units$GM
LB TON MT KG GM
2.20462e-03 1.10230e-06 1.00000e-06 1.00000e-03 1.00000e+00
$mass$mode
[1] "double"
$volume
$volume$units
$volume$units$GAL
GAL L
1.00000 3.78541
$volume$units$L
GAL L
0.264172 1.000000
$volume$mode
[1] "double"
$time
$time$units
$time$units$YR
YR DAY HR MIN SEC
1 365 8760 525600 31540000
$time$units$DAY
YR DAY HR MIN SEC
2.73973e-03 1.00000e+00 2.40000e+01 1.44000e+03 8.64000e+04
$time$units$HR
YR DAY HR MIN SEC
1.14155e-04 4.16667e-02 1.00000e+00 6.00000e+01 3.60000e+03
$time$units$MIN
YR DAY HR MIN SEC
1.90260e-06 6.94444e-04 1.66667e-02 1.00000e+00 6.00000e+01
$time$units$SEC
YR DAY HR MIN SEC
3.17100e-08 1.15740e-05 2.77778e-04 1.66667e-02 1.00000e+00
$time$mode
[1] "double"
$energy
$energy$units
$energy$units$KWH
KWH MJ GGE
1.00000000 3.60000000 0.02967846
$energy$units$MJ
KWH MJ GGE
0.277778000 1.000000000 0.008244023
$energy$units$GGE
KWH MJ GGE
33.69447 121.30000 1.00000
$energy$mode
[1] "double"
$people
$people$units
$people$units$PRSN
PRSN
1
$people$mode
[1] "integer"
$vehicles
$vehicles$units
$vehicles$units$VEH
VEH
1
$vehicles$mode
[1] "integer"
$trips
$trips$units
$trips$units$TRIP
TRIP
1
$trips$mode
[1] "integer"
$households
$households$units
$households$units$HH
HH
1
$households$mode
[1] "integer"
$employment
$employment$units
$employment$units$JOB
JOB
1
$employment$mode
[1] "integer"
$activity
$activity$units
$activity$units$HHJOB
HHJOB
1
10.6.4 Appendix D: Scenario Input File Examples
Figure B1. Example of input file to be loaded into ‘Global’ group
NOTE: Heavy lines denote rows that are hidden to shorten the display
Figure B2. Example of input file to be loaded into ‘forecast year’ group
10.6.5 Appendix E: Example Module Script from the VETransportSupply Package
#======================
#AssignTransitService.R
#======================
#<doc>
#
## AssignTransitService Module
#### November 5, 2018
#
#This module assigns transit service level to the metropolitan area (Marea) and neighborhoods (Bzones). Annual revenue-miles (i.e. transit miles in revenue service) by transit mode type are read from an input file. The following 8 modes are recognized:
#* DR = Demand-responsive
#* VP = Vanpool and similar
#* MB = Standard motor bus
#* RB = Bus rapid transit and commuter bus
#* MG = Monorail/automated guideway
#* SR = Streetcar/trolley bus/inclined plain
#* HR = Heavy Rail/Light Rail
#* CR = Commuter Rail/Hybrid Rail/Cable Car/Aerial Tramway
#
#Revenue miles are converted to bus (i.e. MB) equivalents using factors derived from urbanized are data from the National Transit Database (NTD). Bus-equivalent revenue miles are used in models which predict vehicle ownership and household DVMT.
#
#Revenue miles by mode type are also translated (using NTD data) into vehicle miles by 3 vehicle types: van, bus, and rail. Miles by vehicle type are used to calculate public transit energy consumption and emissions.
#
#The module also reads in user supplied data on relative public transit accessibility by Bzone as explained below.
#
### Model Parameter Estimation
#
#Parameters are calculated to convert the revenue miles for each of the 8 recognized public transit modes into bus equivalents, and to convert revenue miles into vehicle miles. Data extracted from the 2015 National Transit Database (NTD) are used to calculate these parameters. The extracted datasets are in the *2015_Service.csv* and *2015_Agency_information.csv* files in the *inst/extdata* directory of this package. These files contain information about transit service and transit service providers located within urbanized areas. Documentation of the data are contained in the accompanying *2015_Service.txt* and *2015_Agency_information.txt* files.
#
#Bus equivalent factors for each of the 8 modes is calculated on the basis of the average productivity of each mode as measured by the ratio of passenger miles to revenue miles. The bus-equivalency factor of each mode is the ratio of the average productivity of the mode to the average productivity of the bus (MB) mode.
#
#Factors to compute vehicle miles by mode from revenue miles by mode are calculated from the NTD data on revenue miles and deadhead (i.e. out of service) miles. The vehicle mile factor is the sum of revenue and deadhead miles divided by the revenue miles. These factors vary by mode.
#
### How the Module Work
#
#The user supplies data on the annual revenue miles of service by each of the 8 transit modes for the Marea. These revenue miles are converted to bus equivalents using the estimated bus-equivalency factors and summed to calculate total bus-equivalent revenue miles. This value is divided by the urbanized area population of the Marea to compute bus-equivalent revenue miles per capita. This public transit service measure is used in models of household vehicle ownership and household vehicle travel.
#
#The user supplied revenue miles by mode are translated into vehicle miles by mode using the estimated conversion factors. The results are then simplified into 3 vehicle types (Van, Bus, Rail) where the DR and VP modes are assumed to be served by vans, the MB and RB modes are assumed to be served by buses, and the MG, SR, HR, and CR modes are assumed to be served by rail.
#
#The user also supplies information on the aggregate frequency of peak period transit service within 0.25 miles of the Bzone boundary per hour during evening peak period. This is the *D4c* measure included in the Environmental Protection Agency's (EPA) [Smart Location Database] (https://www.epa.gov/smartgrowth/smart-location-database-technical-documentation-and-user-guide). Following is the description of the measure from the user guide:
#>EPA analyzed GTFS data to calculate the frequency of service for each transit route between 4:00 and 7:00 PM on a weekday. Then, for each block group, EPA identified transit routes with service that stops within 0.4 km (0.25 miles). Finally EPA summed total aggregate service frequency by block group. Values for this metric are expressed as service frequency per hour of service.
#
#</doc>
#=============================================
#SECTION 1: ESTIMATE AND SAVE MODEL PARAMETERS
#=============================================
#Describe specifications for transit data files
#----------------------------------------------
#Transit agency data
AgencyInp_ls <- items(
item(
NAME =
items("AgencyID",
"PrimaryUZA",
"Population"),
TYPE = "integer",
PROHIBIT = c("NA", "<= 0"),
ISELEMENTOF = "",
UNLIKELY = "",
TOTAL = ""
),
item(
NAME = "UZAName",
TYPE = "character",
PROHIBIT = "",
ISELEMENTOF = "",
UNLIKELY = "",
TOTAL = ""
)
)
#Transit service data
ServiceInp_ls <- items(
item(
NAME =
items("RevenueMiles",
"DeadheadMiles",
"PassengerMiles"),
TYPE = "double",
PROHIBIT = c("< 0"),
ISELEMENTOF = "",
UNLIKELY = "",
TOTAL = ""
),
item(
NAME =
items("AgencyID",
"AgencyName",
"Mode",
"TimePeriod"),
TYPE = "character",
PROHIBIT = "",
ISELEMENTOF = "",
UNLIKELY = "",
TOTAL = ""
)
)
#Define function to estimate public transit model parameters
#-----------------------------------------------------------
#' Estimate public transit model parameters.
#'
#' \code{estimateTransitModel} estimates transit model parameters.
#'
#' This function estimates transit model parameters from 2015 National Transit
#' Database information on transit agencies and service levels. The function
#' calculates factors for converting annual revenue miles by transit mode to
#' total bus-equivalent revenue miles. It also calculates factors to convert
#' revenue miles by mode into vehicle miles by mode.
#'
#' @return A list containing the following elements:
#' BusEquivalents_df: factors to convert revenue miles by mode into bus
#' equivalents,
#' UZABusEqRevMile_df: data on bus equivalent revenue miles by urbanized area,
#' VehMiFactors_df: factors to convert revenue miles by mode into vehicle miles
#' by mode.
#' @name estimateTransitModel
#' @import stats
#' @export
estimateTransitModel <- function() {
#Read in and process transit datasets
#------------------------------------
#Read in transit agency datasets
Agency_df <-
processEstimationInputs(
AgencyInp_ls,
"2015_Agency_information.csv",
"AssignTransitService.R")
#Read in transit service datasets
Service_df <-
processEstimationInputs(
ServiceInp_ls,
"2015_Service.csv",
"AssignTransitService.R")
#Select only rows with annual totals
Service_df <- Service_df[Service_df$TimePeriod == "Annual Total",]
#Select only rows for service in urbanized areas
Service_df <- Service_df[Service_df$AgencyID %in% Agency_df$AgencyID,]
#Define combined modes and create index datasets
#-----------------------------------------------
CombinedCode_ls <-
list(
DR = c("DR", "DT"),
VP = c("VP", "PB"),
MB = c("MB"),
RB = c("RB", "CB"),
MG = c("MG"),
SR = c("SR", "TB", "IP"),
HR = c("LR", "HR", "AR"),
CR = c("CR", "YR", "CC", "TR")
)
CombinedCode_ <-
c(DR = "DR", DT = "DR", VP = "VP", PB = "VP", MB = "MB", RB = "RB", CB = "RB",
MG = "MG", SR = "SR", TB = "SR", IP = "SR", LR = "HR", HR = "HR", AR = "HR",
CR = "CR", YR = "CR", CC = "CR", TR = "CR"
)
Cm <- c("DR", "VP", "MB", "RB", "MG", "SR", "HR", "CR")
#Calculate bus equivalency factors
#---------------------------------
#Calculate productivity measure
Service_df$Productivity <-
Service_df$PassengerMiles / Service_df$RevenueMiles
#Calculate the average productivity by mode
AveProductivity_Md <-
tapply(Service_df$Productivity, Service_df$Mode, mean, na.rm = TRUE)
#Calculate bus equivalency of different modes
BusEquiv_Md <- AveProductivity_Md / AveProductivity_Md["MB"]
#Calculate average productivity by combined mode
BusEquiv_Cm <- unlist(lapply(CombinedCode_ls, function(x) {
mean(BusEquiv_Md[x])
}))
#Create data frame with mode names and equivalency factors
BusEquiv_df <-
data.frame(
Mode = names(BusEquiv_Cm),
BusEquivalents = unname(BusEquiv_Cm)
)
#Calculate revenue miles to total vehicle mile factors by mode
#-------------------------------------------------------------
#Convert DeadheadMiles for mode DT from NA to 0
Service_df$DeadheadMiles[Service_df$Mode == "DT"] <- 0
#Create data frame of complete cases of revenue miles and deadhead miles
Veh_df <- Service_df[, c("Mode", "RevenueMiles", "DeadheadMiles")]
Veh_df <- Veh_df[complete.cases(Veh_df),]
#Calculate total revenue miles by combined mode
RevMi_Md <- tapply(Veh_df$RevenueMiles, Veh_df$Mode, sum)
RevMi_Cm <- unlist(lapply(CombinedCode_ls, function(x) {
sum(RevMi_Md[x])
}))
#Calculate total deadhead miles by combined mode
DeadMi_Md <- tapply(Veh_df$DeadheadMiles, Veh_df$Mode, sum)
DeadMi_Cm <- unlist(lapply(CombinedCode_ls, function(x) {
sum(DeadMi_Md[x])
}))
#Calculate vehicle mile factors by combined mode
VehMiFactors_Cm <- (RevMi_Cm + DeadMi_Cm) / RevMi_Cm
VehMiFactors_df <-
data.frame(
Mode = names(VehMiFactors_Cm),
VehMiFactors = unname(VehMiFactors_Cm)
)
#Calculate bus equivalent transit service by urbanized area
#----------------------------------------------------------
#Attach urbanized area code to service data
Service_df$UzaCode <- Agency_df$PrimaryUZA[match(Service_df$AgencyID, Agency_df$AgencyID)]
Service_df$UzaName <- Agency_df$UZAName[match(Service_df$AgencyID, Agency_df$AgencyID)]
#Tabulate vehicle revenue miles by urbanized area and mode
RevMi_UnMd <-
tapply(Service_df$RevenueMiles,
list(Service_df$UzaName, Service_df$Mode),
sum)
RevMi_UnMd[is.na(RevMi_UnMd)] <- 0
#Summarize by combined mode
RevMi_UnCm <- t(apply(RevMi_UnMd, 1, function(x) {
tapply(x, CombinedCode_[colnames(RevMi_UnMd)], sum, na.rm = TRUE)[Cm]
}))
#Sum up the bus-equivalent revenue miles by urbanized area
BusEqRevMi_Un <-
rowSums(sweep(RevMi_UnCm, 2, BusEquiv_Cm, "*"))
#Tabulate population by urbanized area
UzaPop_Un <- Agency_df$Population[!duplicated(Agency_df$PrimaryUZA)]
names(UzaPop_Un) <- Agency_df$UZAName[!duplicated(Agency_df$PrimaryUZA)]
UzaPop_Un <- UzaPop_Un[names(BusEqRevMi_Un)]
UzaPop_Un <- UzaPop_Un[names(BusEqRevMi_Un)]
#Calculate bus-equivalent revenue miles per capita
BusEqRevMiPC_Un <- BusEqRevMi_Un / UzaPop_Un
#Create data frame of urbanized area bus revenue mile equivalency
UZABusEqRevMile_df <-
Service_df[!duplicated(Service_df$UzaName), c("UzaCode", "UzaName")]
rownames(UZABusEqRevMile_df) <- UZABusEqRevMile_df$UzaName
UZABusEqRevMile_df <- UZABusEqRevMile_df[names(BusEqRevMi_Un),]
UZABusEqRevMile_df$BusEqRevMi <- unname(BusEqRevMi_Un)
UZABusEqRevMile_df$UzaPop <- unname(UzaPop_Un)
UZABusEqRevMile_df$BusEqRevMiPC <- unname(BusEqRevMiPC_Un)
rownames(UZABusEqRevMile_df) <- NULL
#Return the results
#------------------
list(
BusEquivalents_df = BusEquiv_df,
UZABusEqRevMile_df = UZABusEqRevMile_df,
VehMiFactors_df = VehMiFactors_df
)
}
#Estimate public transit model parameters
#----------------------------------------
TransitParam_ls <- estimateTransitModel()
BusEquivalents_df <- TransitParam_ls$BusEquivalents_df
UZABusEqRevMile_df <- TransitParam_ls$UZABusEqRevMile_df
VehMiFactors_df <- TransitParam_ls$VehMiFactors_df
rm(AgencyInp_ls)
rm(ServiceInp_ls)
#Save the bus equivalency factors
#--------------------------------
#' Bus equivalency factors
#'
#' Bus revenue mile equivalency factors to convert revenue miles for various
#' modes to bus-equivalent revenue miles.
#'
#' @format A data frame with 8 rows and 2 variables containing factors for
#' converting revenue miles of various modes to bus equivalent revenue miles.
#' Mode names are 2-character codes corresponding to consolidated mode types.
#' Consolidated mode types represent modes that have similar characteristics and
#' bus equivalency values. The consolidate mode codes and their meanings are as
#' follows:
#' DR = Demand-responsive
#' VP = Vanpool and similar
#' MB = Standard motor bus
#' RB = Bus rapid transit and commuter bus
#' MG = Monorail/automated guideway
#' SR = Streetcar/trolley bus/inclined plain
#' HR = Heavy Rail/Light Rail
#' CR = Commuter Rail/Hybrid Rail/Cable Car/Aerial Tramway
#'
#' \describe{
#' \item{Mode}{abbreviation for consolidated mode}
#' \item{BusEquivalents}{numeric factor for converting revenue miles to bus equivalents}
#' }
#' @source AssignTransitService.R script.
"BusEquivalents_df"
usethis::use_data(BusEquivalents_df, overwrite = TRUE)
#Save the vehicle mile factors
#-----------------------------
#' Revenue miles to vehicle miles conversion factors
#'
#' Vehicle mile factors convert revenue miles for various modes to vehicle
#' miles for those modes.
#'
#' @format A data frame with 8 rows and 2 variables containing factors for
#' converting revenue miles of various modes to vehicle miles.
#' Mode names are 2-character codes corresponding to consolidated mode types.
#' Consolidated mode types represent modes that have similar characteristics and
#' bus equivalency values. The consolidate mode codes and their meanings are as
#' follows:
#' DR = Demand-responsive
#' VP = Vanpool and similar
#' MB = Standard motor bus
#' RB = Bus rapid transit and commuter bus
#' MG = Monorail/automated guideway
#' SR = Streetcar/trolley bus/inclined plain
#' HR = Heavy Rail/Light Rail
#' CR = Commuter Rail/Hybrid Rail/Cable Car/Aerial Tramway
#'
#' \describe{
#' \item{Mode}{abbreviation for consolidated mode}
#' \item{VehMiFactors}{numeric factors for converting revenue miles to
#' vehicle miles}
#' }
#' @source AssignTransitService.R script.
"VehMiFactors_df"
usethis::use_data(VehMiFactors_df, overwrite = TRUE)
#Save the urbanized area bus equivalency data
#--------------------------------------------
#' Urbanized area bus equivalent revenue mile data for 2015
#'
#' Urbanized area data from the 2015 National Transit Database (NTD) related to
#' the calculation of bus equivalent revenue miles and per capita values.
#'
#' @format A data frame with 439 rows and 5 variables containing urbanized area
#' data on bus equivalent revenue miles
#'
#' \describe{
#' \item{UzaCode}{integer code corresponding to 5-digit code used in the NTD}
#' \item{UzaName}{urbanized area name}
#' \item{BusEqRevMi}{annual bus equivalent revenue miles in the urbanized area}
#' \item{UzaPop}{urbanized area population}
#' \item{BusEqRevMiPC}{annual bus equivalent revenue miles per capita in the urbanized area}
#' }
#' @source AssignTransitService.R script.
"UZABusEqRevMile_df"
usethis::use_data(UZABusEqRevMile_df, overwrite = TRUE)
#Clean up
rm(TransitParam_ls)
#================================================
#SECTION 2: DEFINE THE MODULE DATA SPECIFICATIONS
#================================================
#Define the data specifications
#------------------------------
AssignTransitServiceSpecifications <- list(
#Level of geography module is applied at
RunBy = "Region",
#Specify new tables to be created by Inp if any
#Specify new tables to be created by Set if any
#Specify input data
Inp = items(
item(
NAME =
items(
"DRRevMi",
"VPRevMi",
"MBRevMi",
"RBRevMi",
"MGRevMi",
"SRRevMi",
"HRRevMi",
"CRRevMi"),
FILE = "marea_transit_service.csv",
TABLE = "Marea",
GROUP = "Year",
TYPE = "compound",
UNITS = "MI/YR",
NAVALUE = -1,
SIZE = 0,
PROHIBIT = c("NA", "< 0"),
ISELEMENTOF = "",
UNLIKELY = "",
TOTAL = "",
DESCRIPTION =
list(
"Annual revenue-miles of demand-responsive public transit service",
"Annual revenue-miles of van-pool and similar public transit service",
"Annual revenue-miles of standard bus public transit service",
"Annual revenue-miles of rapid-bus and commuter bus public transit service",
"Annual revenue-miles of monorail and automated guideway public transit service",
"Annual revenue-miles of streetcar and trolleybus public transit service",
"Annual revenue-miles of light rail and heavy rail public transit service",
"Annual revenue-miles of commuter rail, hybrid rail, cable car, and aerial tramway public transit service"
)
),
item(
NAME = "D4c",
FILE = "bzone_transit_service.csv",
TABLE = "Bzone",
GROUP = "Year",
TYPE = "double",
UNITS = "aggregate peak period transit service",
NAVALUE = -1,
SIZE = 0,
PROHIBIT = c("NA", "< 0"),
ISELEMENTOF = "",
UNLIKELY = "",
TOTAL = "",
DESCRIPTION = "Aggregate frequency of transit service within 0.25 miles of block group boundary per hour during evening peak period (Ref: EPA 2010 Smart Location Database)"
)
),
#Specify data to be loaded from data store
Get = items(
item(
NAME = "Marea",
TABLE = "Marea",
GROUP = "Year",
TYPE = "character",
UNITS = "ID",
PROHIBIT = "",
ISELEMENTOF = ""
),
item(
NAME = "Marea",
TABLE = "Bzone",
GROUP = "Year",
TYPE = "character",
UNITS = "ID",
PROHIBIT = "",
ISELEMENTOF = ""
),
item(
NAME =
items(
"DRRevMi",
"VPRevMi",
"MBRevMi",
"RBRevMi",
"MGRevMi",
"SRRevMi",
"HRRevMi",
"CRRevMi"),
TABLE = "Marea",
GROUP = "Year",
TYPE = "compound",
UNITS = "MI/YR",
PROHIBIT = c("NA", "< 0"),
ISELEMENTOF = ""
),
item(
NAME = "UrbanPop",
TABLE = "Bzone",
GROUP = "Year",
TYPE = "people",
UNITS = "PRSN",
PROHIBIT = c("NA", "< 0"),
ISELEMENTOF = ""
)
),
#Specify data to saved in the data store
Set = items(
item(
NAME = "TranRevMiPC",
TABLE = "Marea",
GROUP = "Year",
TYPE = "compound",
UNITS = "MI/PRSN/YR",
NAVALUE = -1,
PROHIBIT = c("NA", "< 0"),
ISELEMENTOF = "",
SIZE = 0,
DESCRIPTION = "Ratio of annual bus-equivalent revenue-miles (i.e. revenue-miles at the same productivity - passenger miles per revenue mile - as standard bus) to urbanized area population"
),
item(
NAME =
items(
"VanDvmt",
"BusDvmt",
"RailDvmt"
),
TABLE = "Marea",
GROUP = "Year",
TYPE = "compound",
UNITS = "MI/DAY",
NAVALUE = -1,
PROHIBIT = c("NA", "< 0"),
ISELEMENTOF = "",
SIZE = 0,
DESCRIPTION = items(
"Total daily miles traveled by vans of various sizes to provide demand responsive, vanpool, and similar services.",
"Total daily miles traveled by buses of various sizes to provide bus service of various types.",
"Total daily miles traveled by light rail, heavy rail, commuter rail, and similar types of vehicles."
)
)
)
)
#Save the data specifications list
#---------------------------------
#' Specifications list for AssignTransitService module
#'
#' A list containing specifications for the AssignTransitService module.
#'
#' @format A list containing 4 components:
#' \describe{
#' \item{RunBy}{the level of geography that the module is run at}
#' \item{Inp}{scenario input data to be loaded into the datastore for this
#' module}
#' \item{Get}{module inputs to be read from the datastore}
#' \item{Set}{module outputs to be written to the datastore}
#' }
#' @source AssignTransitService.R script.
"AssignTransitServiceSpecifications"
usethis::use_data(AssignTransitServiceSpecifications, overwrite = TRUE)
#=======================================================
#SECTION 3: DEFINE FUNCTIONS THAT IMPLEMENT THE SUBMODEL
#=======================================================
#This function calculates the annual bus equivalent revenue miles per capita for
#the urbanized area from the number of annual revenue miles for different
#public transit modes and the urban area population.
#Main module function that calculates bus equivalent revenue miles per capita
#----------------------------------------------------------------------------
#' Calculate bus equivalent revenue miles per capita by Marea.
#'
#' \code{AssignTransitService} calculate bus equivalent revenue miles per capita.
#'
#' This function calculates bus equivalent revenue miles per capita for each
#' Marea.
#'
#' @param L A list containing the components listed in the Get specifications
#' for the module.
#' @return A list containing the components specified in the Set
#' specifications for the module.
#' @name AssignTransitService
#' @import visioneval
#' @export
AssignTransitService <- function(L) {
#Set up
#------
#Fix seed as synthesis involves sampling
set.seed(L$G$Seed)
#Define vector of modes
Md <- as.character(BusEquivalents_df$Mode)
#Define vector of Mareas
Ma <- L$Year$Marea$Marea
#Calculate bus equivalent revenue miles
#--------------------------------------
#Make table of revenue miles by Marea
RevMi_df <- data.frame(L$Year$Marea[paste0(Md, "RevMi")])
colnames(RevMi_df) <- Md
rownames(RevMi_df) <- Ma
RevMi_MaMd <- as.matrix(RevMi_df)
#Calculate the bus equivalent revenue miles
BusEq_Md <- BusEquivalents_df$BusEquivalents
names(BusEq_Md) <- Md
BusEqRevMi_Ma <-
rowSums(sweep(RevMi_MaMd, 2, BusEq_Md, "*"))[Ma]
#Calculate the bus equivalent revenue miles per capita
#-----------------------------------------------------
#Calculate population in the urbanized area
UrbanPop_Ma <-
tapply(L$Year$Bzone$UrbanPop, L$Year$Bzone$Marea, sum)[Ma]
#Calculate Marea bus equivalent revenue miles per capita
TranRevMiPC_Ma <- BusEqRevMi_Ma / UrbanPop_Ma
#Calculate vehicle miles by vehicle type
#---------------------------------------
#Make vector of vehicle miles factors conforming with RevMi_df
VehMiFactors_Md <- VehMiFactors_df$VehMiFactors
names(VehMiFactors_Md) <- VehMiFactors_df$Mode
VehMiFactors_Md <- VehMiFactors_Md[names(RevMi_df)]
#Calculate daily vehicle miles by Marea and mode
VehMi_MaMd <- as.matrix(sweep(RevMi_df, 2, VehMiFactors_Md, "*")) / 365
#Define correspondence between modes and vehicle types
ModeToVehType_ <- c(
DR = "Van",
VP = "Van",
MB = "Bus",
RB = "Bus",
MG = "Rail",
SR = "Rail",
HR = "Rail",
CR = "Rail"
)
ModeToVehType_ <- ModeToVehType_[colnames(VehMi_MaMd)]
VehMi_df <-
data.frame(
t(
apply(VehMi_MaMd, 1, function(x) {
tapply(x, ModeToVehType_, sum) })
)
)
#Return the results
#------------------
#Initialize output list
Out_ls <- initDataList()
Out_ls$Year$Marea <-
list(TranRevMiPC = TranRevMiPC_Ma,
VanDvmt = VehMi_df$Van,
BusDvmt = VehMi_df$Bus,
RailDvmt = VehMi_df$Rail)
#Return the outputs list
Out_ls
}
#===============================================================
#SECTION 4: MODULE DOCUMENTATION AND AUXILLIARY DEVELOPMENT CODE
#===============================================================
#Run module automatic documentation
#----------------------------------
documentModule("AssignTransitService")
#Test code to check specifications, loading inputs, and whether datastore
#contains data needed to run module. Return input list (L) to use for developing
#module functions
#-------------------------------------------------------------------------------
# library(filesstrings)
# library(visioneval)
# source("tests/scripts/test_functions.R")
# #Set up test environment
# TestSetup_ls <- list(
# TestDataRepo = "../Test_Data/VE-RSPM",
# DatastoreName = "Datastore.tar",
# LoadDatastore = TRUE,
# TestDocsDir = "verspm",
# ClearLogs = TRUE,
# # SaveDatastore = TRUE
# SaveDatastore = FALSE
# )
# setUpTests(TestSetup_ls)
# #Run test module
# TestDat_ <- testModule(
# ModuleName = "AssignTransitService",
# LoadDatastore = TRUE,
# SaveDatastore = TRUE,
# DoRun = FALSE
# )
# L <- TestDat_$L
# R <- AssignTransitService(L)10.6.6 Appendix F: Example Test Script from the VESimHouseholds Package
#vestate_test.R
#--------------
#Load packages and test functions
library(visioneval)
library(filesstrings)
source("tests/scripts/test_functions.R")
#Define test setup parameters
TestSetup_ls <- list(
TestDataRepo = "../Test_Data/VE-State",
DatastoreName = "Datastore.tar",
LoadDatastore = FALSE,
TestDocsDir = "vestate",
ClearLogs = TRUE,
# SaveDatastore = TRUE
SaveDatastore = FALSE
)
#Define the module tests
Tests_ls <- list(
list(ModuleName = "CreateHouseholds", LoadDatastore = FALSE, SaveDatastore = TRUE, DoRun = TRUE),
list(ModuleName = "PredictWorkers", LoadDatastore = TRUE, SaveDatastore = TRUE, DoRun = TRUE),
list(ModuleName = "AssignLifeCycle", LoadDatastore = TRUE, SaveDatastore = TRUE, DoRun = TRUE),
list(ModuleName = "PredictIncome", LoadDatastore = TRUE, SaveDatastore = TRUE, DoRun = TRUE)
)
#Set up, run tests, and save test results
setUpTests(TestSetup_ls)
doTests(Tests_ls, TestSetup_ls)
saveTestResults(TestSetup_ls)
10.6.8 getYears: Retrieve years
10.6.8.1 Description
getYears a visioneval framework model user function that reads the
Years component from the the model state file.
10.6.8.3 Details
This is a convenience function to make it easier to retrieve the Years component of the model state file which lists all of the specified model run years. If the Years component includes the base year, then the returned vector of years places the base year first in the order. This ordering is important because some modules calculate future year values by pivoting off of base year values so the base year must be run first.
10.6.9 initializeModel: Initialize model.
10.6.9.1 Description
initializeModel a visioneval framework model user function
that initializes a VisionEval model, loading all parameters and inputs, and
making checks to ensure that model can run successfully.
10.6.9.2 Usage
initializeModel(ParamDir = "defs",
RunParamFile = "run_parameters.json", GeoFile = "geo.csv",
ModelParamFile = "model_parameters.json", LoadDatastore = FALSE,
DatastoreName = NULL, SaveDatastore = TRUE)10.6.9.3 Arguments
| Argument | Description |
|---|---|
ParamDir |
A string identifying the relative or absolute path to the directory where the parameter and geography definition files are located. The default value is “defs.” |
RunParamFile |
A string identifying the name of a JSON-formatted text file that contains parameters needed to identify and manage the model run. The default value is “run_parameters.json.” |
GeoFile |
A string identifying the name of a text file in comma-separated values format that contains the geographic specifications for the model. The default value is “geo.csv.” |
ModelParamFile |
A string identifying the name of a JSON-formatted text file that contains global model parameters that are important to a model and may be shared by several modules. |
LoadDatastore |
A logical identifying whether an existing datastore should be loaded. |
DatastoreName |
A string identifying the full path name of a datastore to load or NULL if an existing datastore in the working directory is to be loaded. |
SaveDatastore |
A string identifying whether if an existing datastore in the working directory should be saved rather than removed. |
10.6.9.4 Details
This function does several things to initialize the model environment and datastore including: 1) Initializing a file that is used to keep track of the state of key model run variables and the datastore; 2) Initializes a log to which messages are written; 3) Creates the datastore and initializes its structure, reads in and checks the geographic specifications and initializes the geography in the datastore, or loads an existing datastore if one has been identified; 4) Parses the model run script to identify the modules in their order of execution and checks whether all the identified packages are installed and the modules exist in the packages; 5) Checks that all data requested from the datastore will be available when it is requested and that the request specifications match the datastore specifications; 6) Checks all of the model input files to determine whether they they are complete and comply with specifications.
10.6.9.5 Value
None. The function prints to the log file messages which identify whether or not there are errors in initialization. It also prints a success message if initialization has been successful.
10.6.9.6 Calls
assignDatastoreFunctions, checkDataset, checkModuleExists, checkModuleSpecs, getModelState, getModuleSpecs, initDatastoreGeography, initLog, initModelStateFile, inputsToDatastore, loadDatastore, loadModelParameters, parseModelScript, processModuleInputs, processModuleSpecs, readGeography, readModelState, setModelState, simDataTransactions, writeLog
10.6.10 readDatastoreTables: Read multiple datasets from multiple tables in datastores
10.6.10.1 Description
readDatastoreTables a visioneval framework model user function that
reads datasets from one or more tables in a specified group in one or more
datastores
10.6.10.3 Arguments
| Argument | Description |
|---|---|
Tables_ls |
a named list where the name of each component is the name of a table in a datastore group and the value is a string vector of the names of the datasets to be retrieved. |
Group |
a string that is the name of the group to retrieve the table datasets from. |
DstoreLocs_ |
a string vector identifying the paths to all of the datastores to extract the datasets from. Each entry must be the full relative path to a datastore (e.g. ‘tests/Datastore’). |
DstoreType |
a string identifying the type of datastore (e.g. ‘RD,’ ‘H5’). Note |
10.6.10.4 Details
This function can read multiple datasets in one or more tables in a group. More than one datastore my be specified so that if datastore references are used in a model run, datasets from the referenced datastores may be queried as well. Note that the capability for querying multiple datastores is only for the purpose of querying datastores for a single model scenario. This capability should not be used to compare multiple scenarios. The function does not segregate datasets by datastore. Attempting to use this function to compare multiple scenarios could produce unpredictable results.
10.6.10.5 Value
A named list having two components. The ‘Data’ component is a list containing the datasets from the datastores where the name of each component of the list is the name of a table from which identified datasets are retrieved and the value is a data frame containing the identified datasets. The ‘Missing’ component is a list which identifies the datasets that are missing in each table.
10.6.11 runModule: Run module.
10.6.11.3 Arguments
| Argument | Description |
|---|---|
ModuleName |
A string identifying the name of a module object. |
PackageName |
A string identifying the name of the package the module is a part of. |
RunFor |
A string identifying whether to run the module for all years “AllYears,” only the base year “BaseYear,” or for all years except the base year “NotBaseYear.” |
RunYear |
A string identifying the run year. |
StopOnErr |
A logical identifying whether model execution should be stopped if the module transmits one or more error messages or whether execution should continue with the next module. The default value is TRUE. This is how error handling will ordinarily proceed during a model run. A value of FALSE is used when ‘Initialize’ modules in packages are run during model initialization. These ‘Initialize’ modules are used to check and preprocess inputs. For this purpose, the module will identify any errors in the input data, the ‘initializeModel’ function will collate all the data errors and print them to the log. |
10.6.13 addErrorMsg: Add an error message to the results list
10.6.13.1 Description
addErrorMsg a visioneval framework module developer function that adds
an error message to the Errors component of the module results list that is
passed back to the framework.
10.6.13.3 Arguments
| Argument | Description |
|---|---|
ResultsListName |
the name of the results list given as a character string |
ErrMsg |
a character string that contains the error message |
10.6.13.4 Details
This function is a convenience function for module developers for passing error messages back to the framework. The preferred method for handling errors in module execution is for the module to handle the error by passing one or more error messages back to the framework. The framework will then write error messages to the log and stop execution. Error messages are stored in a component of the returned list called Errors. This component is a string vector where each element is an error message. The addErrorMsg will create the Error component if it does not already exist and will add an error message to the vector.
10.6.14 addWarningMsg: Add a warning message to the results list
10.6.14.1 Description
addWarningMsg a visioneval framework module developer function that
adds an warning message to the Warnings component of the module results list
that is passed back to the framework.
10.6.14.3 Arguments
| Argument | Description |
|---|---|
ResultsListName |
the name of the results list given as a character string |
WarnMsg |
a character string that contains the warning message |
10.6.14.4 Details
This function is a convenience function for module developers for passing warning messages back to the framework. The preferred method for handling warnings in module execution is for the module to handle the warning by passing one or more warning messages back to the framework. The framework will then write warning messages to the log and stop execution. Warning messages are stored in a component of the returned list called Warnings. This component is a string vector where each element is an warning message. The addWarningMsg will create the Warning component if it does not already exist and will add a warning message to the vector.
10.6.15 applyBinomialModel: Applies an estimated binomial model to a set of input values.
10.6.15.1 Description
applyBinomialModel a visioneval framework module developer function
that applies an estimated binomial model to a set of input data.
10.6.15.2 Usage
applyBinomialModel(Model_ls, Data_df, TargetProp = NULL,
CheckTargetSearchRange = FALSE, ApplyRandom = TRUE,
ReturnProbs = FALSE)10.6.15.3 Arguments
| Argument | Description |
|---|---|
Model_ls |
a list which contains the following components: ‘Type’ which has a value of ‘binomial’; ‘Formula’ a string representation of the model equation; ‘Choices’ a two-element vector listing the choice set. The first element is the choice that the binary logit model equation predicts the odds of; ‘PrepFun’ a function which prepares the input data frame for the model application. If no preparation, this element of the list should not be present or should be set equal to NULL; ‘SearchRange’ a two-element numeric vector which specifies the acceptable search range to use when determining the factor for adjusting the model constant. ‘RepeatVar’ a string which identifies the name of a field to use for repeated draws of the model. This is used in the case where for example the input data is households and the output is vehicles and the repeat variable is the number of vehicles in the household. ‘ApplyRandom’ a logical identifying whether the results will be affected by random draws (i.e. if a random number in range 0 - 1 is less than the computed probability) or if a probability cutoff is used (i.e. if the computed probability is greater then 0.5). This is an optional component. If it isn’t present, the function runs with ApplyRandom = TRUE. |
Data_df |
a data frame containing the data required for applying the model. |
TargetProp |
a number identifying a target proportion for the default choice to be achieved for the input data or NULL if there is no target proportion to be achieved. |
CheckTargetSearchRange |
a logical identifying whether the function is to only check whether the specified ‘SearchRange’ for the model will produce acceptable values (i.e. no NA or NaN values). If FALSE (the default), the function will run the model and will not check the target search range. |
ApplyRandom |
a logical identifying whether the outcome will be be affected by random draws (i.e. if a random number in range 0 - 1 is less than the computed probability) or if a probability cutoff is used (i.e. if the computed probability is greater than 0.5) |
ReturnProbs |
a logical identifying whether to return the calculated probabilities rather than the assigned results. The default value is FALSE. |
10.6.15.4 Details
The function calculates the result of applying a binomial logit model to a set of input data. If a target proportion (TargetProp) is specified, the function calls the ‘binarySearch’ function to calculate an adjustment to the constant of the model equation so that the population proportion matches the target proportion. The function will also test whether the target search range specified for the model will produce acceptable values.
10.6.16 applyLinearModel: Applies an estimated linear model to a set of input values.
10.6.16.1 Description
applyLinearModel a visioneval framework module developer function that
applies an estimated linear model to a set of input data.
10.6.16.2 Usage
applyLinearModel(Model_ls, Data_df, TargetMean = NULL,
CheckTargetSearchRange = FALSE)10.6.16.3 Arguments
| Argument | Description |
|---|---|
Model_ls |
a list which contains the following components: ‘Type’ which has a value of ‘linear’; ‘Formula’ a string representation of the model equation; ‘PrepFun’ a function which prepares the input data frame for the model application. If no preparation, this element of the list should not be present or should be set equal to NULL; ‘SearchRange’ a two-element numeric vector which specifies the acceptable search range to use when determining the dispersion factor. ‘OutFun’ a function that is applied to transform the results of applying the linear model. For example to untransform a power-transformed variable. If no transformation is necessary, this element of the list should not be present or should be set equal to NULL. |
Data_df |
a data frame containing the data required for applying the model. |
TargetMean |
a number identifying a target mean value to be achieved or NULL if there is no target. |
CheckTargetSearchRange |
a logical identifying whether the function is to only check whether the specified ‘SearchRange’ for the model will produce acceptable values (i.e. no NA or NaN values). If FALSE (the default), the function will run the model and will not check the target search range. |
10.6.16.4 Details
The function calculates the result of applying a linear regression model to a set of input data. If a target mean value (TargetMean) is specified, the function calculates a standard deviation of a sampling distribution which is applied to linear model results. For each value returned by the linear model, a sample is drawn from a normal distribution where the mean value of the distribution is the linear model result and the standard deviation of the distibution is calculated by the binary search to match the population mean value to the target mean value. This process is meant to be applied to linear model where the dependent variable is power transformed. Applying the sampling distribution to the linear model results increases the dispersion of results to match the observed dispersion and also matches the mean values of the untransformed results. This also enables the model to be applied to situations where the mean value is different than the observed mean value.
10.6.17 binarySearch: Binary search function to find a parameter which achieves a target value.
10.6.17.1 Description
binarySearch a visioneval framework module developer function that
uses a binary search algorithm to find the value of a function parameter for
which the function achieves a target value.
10.6.17.2 Usage
binarySearch(Function, SearchRange_, ..., Target = 0, DoWtAve = TRUE,
MaxIter = 100, Tolerance = 1e-04)10.6.17.3 Arguments
| Argument | Description |
|---|---|
Function |
a function which returns a value which is compared to the ‘Target’ argument. The function must take as its first argument a value which from the ‘SearchRange_.’ It must return a value that may be compared to the ‘Target’ value. |
SearchRange_ |
a two element numeric vector which has the lowest and highest values of the parameter range within which the search will be carried out. |
... |
one or more optional arguments for the ‘Function.’ |
Target |
a numeric value that is compared with the return value of the ‘Function.’ |
DoWtAve |
a logical indicating whether successive weighted averaging is to be done. This is useful for getting stable results for stochastic calculations. |
MaxIter |
an integer specifying the maximum number of iterations to all the search to attempt. |
Tolerance |
a numeric value specifying the proportional difference between the ‘Target’ and the return value of the ‘Function’ to determine when the search is complete. |
10.6.17.4 Details
A binary search algorithm is used by several modules to calibrate the intercept of a binary logit model to match a specified proportion or to calibrate a dispersion parameter for a linear model to match a mean value. This function implements a binary search algorithm in a consistent manner to be used in all modules that need it. It is written to work with stochastic models which by their nature don’t produce the same outputs given the same inputs and so will not converge reliably. To deal with the stochasticity, this function uses a successive averaging approach to smooth out the effect of stochastic variation on reliable convergence. Rather than use the results of a single search iteration to determine the next value range to use in the search, a weighted average of previous values is used with the more recent values being weighted more heavily.
10.6.18 checkModuleOutputs: Check module outputs for consistency with specifications
10.6.18.1 Description
checkModuleOutputs a visioneval framework module developer function
that checks output list produced by a module for consistency with the
module’s specifications.
10.6.18.3 Arguments
| Argument | Description |
|---|---|
Data_ls |
A list of all the datasets returned by a module in the standard list form required by the VisionEval model system. |
ModuleSpec_ls |
A list of module specifications in the standard list form required by the VisionEval model system. |
ModuleName |
A string identifying the name of the module. |
10.6.18.4 Details
This function is used to check whether the output list produced by a module is consistent with the module’s specifications. If there are any specifications for creating tables, the function checks whether the output list contains the table(s), if the LENGTH attribute of the table(s) are present, and if the LENGTH attribute(s) are consistent with the length of the datasets to be saved in the table(s). Each of the datasets in the output list are checked against the specifications. These include checking that the data type is consistent with the specified type and whether all values are consistent with PROHIBIT and ISELEMENTOF conditions. For character types, a check is made to ensure that a SIZE attribute exists and that the size is sufficient to store all characters.
10.6.19 documentModule: Produces markdown documentation for a module
10.6.19.1 Description
documentModule a visioneval framework module developer function
that creates a vignettes directory if one does not exist and produces
module documentation in markdown format which is saved in the vignettes
directory.
10.6.19.3 Arguments
| Argument | Description |
|---|---|
ModuleName |
A string identifying the name of the module (e.g. ‘CalculateHouseholdDvmt’) |
10.6.19.4 Details
This function produces documentation for a module in markdown format. A
‘vignettes’ directory is created if it does not exist and the markdown file
and any associated resources such as image files are saved in that directory.
The function is meant to be called within and at the end of the module
script. The documentation is created from a commented block within the
module script which is enclosed by the opening tag,
10.6.19.5 Value
None. The function has the side effects of creating a ‘vignettes’ directory if one does not exist, copying identified ‘fig’ resources to the ‘vignettes’ directory, and saving the markdown documentation file to the ‘vignettes’ directory. The markdown file is named with the module name and has a ‘md’ suffix.
10.6.20 getRegisteredGetSpecs: Returns Get specifications for registered datasets.
10.6.20.1 Description
getRegisteredGetSpecs a visioneval framework module developer function
that returns a data frame of Get specifications for datasets in the
VisionEval name registry.
10.6.20.3 Arguments
| Argument | Description |
|---|---|
Names_ |
A character vector of the dataset names to get specifications for. |
Tables_ |
A character vector of the tables that the datasets are a part of. |
Groups_ |
A character vector of the groups that the tables are a part of. |
NameRegistryDir |
a string identifying the path to the directory where the name registry file is located. |
10.6.20.4 Details
The VisionEval name registry (VENameRegistry.json) keeps track of the dataset names created by all registered modules by reading in datasets specified in the module Inp specifications or by returning calculated datasets as specified in the module Set specifications. This function reads in the name registry and returns Get specifications for identified datasets.
10.6.21 initDataList: Initialize a list for data transferred to and from datastore
10.6.21.1 Description
initDataList a visioneval framework module developer function that
creates a list to be used for transferring data to and from the datastore.
10.6.21.3 Details
This function initializes a list to store data that is transferred from the datastore to a module or returned from a module to be saved in the datastore. The list has 3 named components (Global, Year, and BaseYear). This is the standard structure for data being passed to and from a module and the datastore.
10.6.22 item: Alias for list function.
10.6.22.1 Description
item a visioneval framework module developer function that is an alias
for the list function whose purpose is to make module specifications easier
to read.
10.6.23 items: Alias for list function.
10.6.23.1 Description
items a visioneval framework module developer function that is
an alias for the list function whose purpose is to make module specifications
easier to read.
10.6.24 loadPackageDataset: Load a VisionEval package dataset
10.6.24.1 Description
loadPackageDataset a visioneval framework module developer function
which loads a dataset identified by name from the VisionEval package
containing the dataset.
10.6.24.4 Details
This function is used to load a dataset identified by name from the VisionEval package which contains the dataset. Using this function is the preferred alternative to hard-wiring the loading using package::dataset notation because it enables users to switch between module versions contained in different packages. For example, there may be different versions of the VEPowertrainsAndFuels package which have different default assumptions about light-duty vehicle powertrain mix and characteristics by model year. Using this function, the module developer only needs to identify the dataset name. The function uses DatasetsByPackage_df data frame in the model state list to identify the package which contains the dataset. It then retrieves and returns the dataset
10.6.25 makeModelFormulaString: Makes a string representation of a model equation.
10.6.25.1 Description
makeModelFormulaString a visioneval framework module developer
function that creates a string equivalent of a model equation.
10.6.25.3 Arguments
| Argument | Description |
|---|---|
EstimatedModel |
the return value of the ‘lm’ or ‘glm’ functions. |
10.6.25.4 Details
The return values of model estimation functions such as ‘lm’ and ‘glm’ contain a large amount of information in addition to the parameter estimates for the specified model. This is particularly the case when the estimation dataset is large. Most of this information is not needed to apply the model and including it can add substantially to the size of a package that includes several estimated models. All that is really needed to implement an estimated model is an equation of the model terms and estimated coefficients. This function creates a string representation of the model equation.
10.6.26 processEstimationInputs: Load estimation data
10.6.26.1 Description
processEstimationInputs a visioneval framework module developer
function that checks whether specified model estimation data meets
specifications and returns the data in a data frame.
10.6.26.3 Arguments
| Argument | Description |
|---|---|
Inp_ls |
A list that describes the specifications for the estimation file. This list must meet the framework standards for specification description. |
FileName |
A string identifying the file name. This is the file name without any path information. The file must located in the “inst/extdata” directory of the package. |
ModuleName |
A string identifying the name of the module the estimation data is being used in. |
10.6.26.4 Details
This function is used to check whether a specified CSV-formatted data file used in model estimation is correctly formatted and contains acceptable values for all the datasets contained within. The function checks whether the specified file exists in the “inst/extdata” directory. If the file does not exist, the function stops and transmits a standard error message that the file does not exist. If the file does exist, the function reads the file into the data frame and then checks whether it contains the specified columns and that the data meets all specifications. If any of the specifications are not met, the function stops and transmits an error message. If there are no data errors the function returns a data frame containing the data in the file.
10.6.27 readVENameRegistry: Reads the VisionEval name registry.
10.6.27.1 Description
readVENameRegistry a visioneval framework module developer function
that reads the VisionEval name registry and returns a list of data frames
containing the Inp and Set specifications.
10.6.27.3 Arguments
| Argument | Description |
|---|---|
NameRegistryDir |
a string identifying the path to the directory where the name registry file is located. |
10.6.27.4 Details
The VisionEval name registry (VENameRegistry.json) keeps track of the dataset names created by all registered modules by reading in datasets specified in the module Inp specifications or by returning calculated datasets as specified in the module Set specifications. This function reads the VisionEval name registry and returns a list of data frames containing the registered Inp and Set specifications.
10.6.28 testModule: Test module
10.6.28.1 Description
testModule a visioneval framework module developer function that sets
up a test environment and tests a module.
10.6.28.2 Usage
testModule(ModuleName, ParamDir = "defs",
RunParamFile = "run_parameters.json", GeoFile = "geo.csv",
ModelParamFile = "model_parameters.json", LoadDatastore = FALSE,
SaveDatastore = TRUE, DoRun = TRUE, RunFor = "AllYears",
StopOnErr = TRUE, RequiredPackages = NULL, TestGeoName = NULL)10.6.28.3 Arguments
| Argument | Description |
|---|---|
ModuleName |
A string identifying the module name. |
ParamDir |
A string identifying the location of the directory where the run parameters, model parameters, and geography definition files are located. The default value is defs. This directory should be located in the tests directory. |
RunParamFile |
A string identifying the name of the run parameters file. The default value is run_parameters.json. |
GeoFile |
A string identifying the name of the file which contains geography definitions. |
ModelParamFile |
A string identifying the name of the file which contains model parameters. The default value is model_parameters.json. |
LoadDatastore |
A logical value identifying whether to load an existing datastore. If TRUE, it loads the datastore whose name is identified in the run_parameters.json file. If FALSE it initializes a new datastore. |
SaveDatastore |
A logical value identifying whether the module outputs will be written to the datastore. If TRUE the module outputs are written to the datastore. If FALSE the outputs are not written to the datastore. |
DoRun |
A logical value identifying whether the module should be run. If FALSE, the function will initialize a datastore, check specifications, and load inputs but will not run the module but will return the list of module specifications. That setting is useful for module development in order to create the all the data needed to assist with module programming. It is used in conjunction with the getFromDatastore function to create the dataset that will be provided by the framework. The default value for this parameter is TRUE. In that case, the module will be run and the results will checked for consistency with the Set specifications. |
RunFor |
A string identifying what years the module is to be tested for. The value must be the same as the value that is used when the module is run in a module. Allowed values are ‘AllYears,’ ‘BaseYear,’ and ‘NotBaseYear.’ |
StopOnErr |
A logical identifying whether model execution should be stopped if the module transmits one or more error messages or whether execution should continue with the next module. The default value is TRUE. This is how error handling will ordinarily proceed during a model run. A value of FALSE is used when ‘Initialize’ modules in packages are run during model initialization. These ‘Initialize’ modules are used to check and preprocess inputs. For this purpose, the module will identify any errors in the input data, the ‘initializeModel’ function will collate all the data errors and print them to the log. |
RequiredPackages |
A character vector identifying any packages that must be installed in order to test the module because the module either has a soft reference to a module in the package (i.e. the Call spec only identifies the name of the module being called) or a soft reference to a dataset in the module (i.e. only identifies the name of the dataset). The default value is NULL. |
TestGeoName |
A character vector identifying the name of the geographic area for which data is to be loaded. This argument has effect only if the DoRun argument is FALSE. It enables the module developer to choose the geographic area data is to be loaded for when developing a module that is run for geography other than the region. For example if a module is run at the Azone level, the user can specify the name of the Azone that data is to be loaded for. If the name is misspecified an error will be flagged. |
10.6.28.4 Details
This function is used to set up a test environment and test a module to check that it can run successfully in the VisionEval model system. The function sets up the test environment by switching to the tests directory and initializing a model state list, a log file, and a datastore. The user may use an existing datastore rather than initialize a new datastore. The use case for loading an existing datastore is where a package contains several modules that run in sequence. The first module would initialize a datastore and then subsequent modules use the datastore that is modified by testing the previous module. When run this way, it is also necessary to set the SaveDatastore argument equal to TRUE so that the module outputs will be saved to the datastore. The function performs several tests including checking whether the module specifications are written properly, whether the the test inputs are correct and complete and can be loaded into the datastore, whether the datastore contains all the module inputs identified in the Get specifications, whether the module will run, and whether all of the outputs meet the module’s Set specifications. The latter check is carried out in large part by the checkModuleOutputs function that is called.
10.6.28.5 Value
If DoRun is FALSE, the return value is a list containing the module specifications. If DoRun is TRUE, there is no return value. The function writes out messages to the console and to the log as the testing proceeds. These messages include the time when each test starts and when it ends. When a key test fails, requiring a fix before other tests can be run, execution stops and an error message is written to the console. Detailed error messages are also written to the log.
10.6.28.6 Calls
assignDatastoreFunctions, checkDataset, checkModuleOutputs, checkModuleSpecs, createGeoIndexList, getFromDatastore, getModelState, getYears, initDatastoreGeography, initLog, initModelStateFile, inputsToDatastore, loadDatastore, loadModelParameters, processModuleInputs, processModuleSpecs, readGeography, readModelState, setInDatastore, setModelState, writeLog
10.6.30 assignDatastoreFunctions: Assign datastore interaction functions
10.6.30.1 Description
assignDatastoreFunctions a visioneval framework control function that
assigns the values of the functions for interacting with the datastore to the
functions for the declared datastore type.
10.6.30.4 Details
The visioneval framework can work with different types of datastores. For example a datastore which stores datasets in an HDF5 file or a datastore which stores datasets as RData files in a directory hierarchy. This function reads the ‘DatastoreType’ parameter from the model state file and then assigns the common datastore interaction functions the values of the functions for the declared datastore type.
10.6.31 checkDataConsistency: Check data consistency with specification
10.6.31.1 Description
checkDataConsistency a visioneval framework control function that
checks whether data to be written to a dataset is consistent with the dataset
attributes.
10.6.31.3 Arguments
| Argument | Description |
|---|---|
DatasetName |
A string identifying the dataset that is being checked. |
Data_ |
A vector of values that may be of type integer, double, character, or logical. |
DstoreAttr_ |
A named list where the components are the attributes of a dataset. |
10.6.31.4 Details
This function compares characteristics of data to be written to a dataset to the dataset attributes to determine whether they are consistent.
10.6.31.5 Value
A list containing two components, Errors and Warnings. If no inconsistencies are found, both components will have zero-length character vectors. If there are one or more inconsistencies, then these components will hold vectors of error and warning messages. Mismatch between UNITS will produce a warning message. All other inconsistencies will produce error messages.
10.6.32 checkDataset: Check dataset existence
10.6.32.1 Description
checkDataset a visioneval framework control function that checks
whether a dataset exists in the datastore and returns a TRUE or FALSE value
with an attribute of the full path to where the dataset should be located in
the datastore.
10.6.32.3 Arguments
| Argument | Description |
|---|---|
Name |
a string identifying the dataset name. |
Table |
a string identifying the table the dataset is a part of. |
Group |
a string or numeric representation of the group the table is a part of. |
DstoreListing_df |
a dataframe which lists the contents of the datastore as contained in the model state file. |
10.6.32.4 Details
This function checks whether a dataset exists. The dataset is identified by its name and the table and group names it is in. If the dataset is not in the datastore, an error is thrown. If it is located in the datastore, the full path name to the dataset is returned.
10.6.33 checkGeography: Check geographic specifications.
10.6.33.1 Description
checkGeography a visioneval framework control function that checks
geographic specifications file for model.
10.6.33.3 Arguments
| Argument | Description |
|---|---|
Directory |
A string identifying the path to the geographic specifications file. |
Filename |
A string identifying the name of the geographic specifications file. |
10.6.33.4 Details
This function reads the file containing geographic specifications for the model and checks the file entries to determine whether they are internally consistent. This function is called by the readGeography function.
10.6.33.5 Value
A list having two components. The first component, ‘Messages,’ contains a string vector of error messages. It has a length of 0 if there are no error messages. The second component, ‘Update,’ is a list of components to update in the model state file. The components of this list include: Geo, a data frame that contains the geographic specifications; BzoneSpecified, a logical identifying whether Bzones are specified; and CzoneSpecified, a logical identifying whether Czones are specified.
10.6.34 checkInputYearGeo: Check years and geography of input file
10.6.34.1 Description
checkInputYearGeo a visioneval framework control function that checks
the ‘Year’ and ‘Geo’ columns of an input file to determine whether they are
complete and have no duplications.
10.6.34.3 Arguments
| Argument | Description |
|---|---|
Year_ |
the vector extract of the ‘Year’ column from the input data. |
Geo_ |
the vector extract of the ‘Geo’ column from the input data. |
Group |
a string identifying the ‘GROUP’ specification for the data sets contained in the input file. |
Table |
a string identifying the ‘TABLE’ specification for the data sets contained in the input file. |
10.6.34.4 Details
This function checks the ‘Year’ and ‘Geo’ columns of an input file to determine whether there are records for all run years specified for the model and for all geographic areas for the level of geography. It also checks for redundant year and geography entries.
10.6.34.5 Value
A list containing the results of the check. The list has two mandatory components and two optional components. ‘CompleteInput’ is a logical that identifies whether records are present for all years and geographic areas. ‘DupInput’ identifies where are any redundant year and geography entries. If ‘CompleteInput’ is FALSE, the list contains a ‘MissingInputs’ component that is a string identifying the missing year and geography records. If ‘DupInput’ is TRUE, the list contains a component that is a string identifying the duplicated year and geography records.
10.6.35 checkIsElementOf: Check if data values are in a specified set of values
10.6.35.1 Description
checkIsElementOf a visioneval framework control function that checks
whether a data vector contains any elements that are not in an allowed set of
values.
10.6.35.3 Arguments
| Argument | Description |
|---|---|
Data_ |
A vector of data of type integer, double, character, or logical. |
SetElements_ |
A vector of allowed values. |
DataName |
A string identifying the field name of the data being compared (used for composing message identifying non-compliant fields). |
10.6.35.4 Details
This function is used to check whether categorical data values are consistent with the defined set of allowed values.
10.6.36 checkMatchConditions: Check values with conditions.
10.6.36.1 Description
checkMatchConditions a visioneval framework control function that
checks whether a data vector contains any elements that match a set of
conditions.
10.6.36.3 Arguments
| Argument | Description |
|---|---|
Data_ |
A vector of data of type integer, double, character, or logical. |
Conditions_ |
A character vector of valid R comparison expressions or an empty vector if there are no conditions. |
DataName |
A string identifying the field name of the data being compared (used for composing message identifying non-compliant fields). |
ConditionType |
A string having a value of either “PROHIBIT” or “UNLIKELY,” the two data specifications which use conditions. |
10.6.36.4 Details
This function checks whether any of the values in a data vector match one or more conditions. The conditions are specified in a character vector where each element is either “NA” (to match for the existence of NA values) or a character representation of a valid R comparison expression for comparing each element with a specified value (e.g. “< 0,” “> 1,” “!= 10”). This function is used both for checking for the presence of prohibited values and for the presence of unlikely values.
10.6.37 checkMatchType: Check data type
10.6.37.1 Description
checkMatchType a visioneval framework control function that checks
whether the data type of a data vector is consistent with specifications.
10.6.37.3 Arguments
| Argument | Description |
|---|---|
Data_ |
A data vector. |
Type |
A string identifying the specified data type. |
DataName |
A string identifying the field name of the data being compared (used for composing message identifying non-compliant fields). |
10.6.37.4 Details
This function checks whether the data type of a data vector is consistent with a specified data type. An error message is generated if data can’t be coerced into the specified data type without the possibility of error or loss of information (e.g. if a double is coerced to an integer). A warning message is generated if the specified type is ‘character’ but the input data type is ‘integer,’ ‘double’ or ‘logical’ since these can be coerced correctly, but that may not be what is intended (e.g. zone names may be input as numbers). Note that some modules may use NA inputs as a flag to identify case when result does not need to match a target. In this case, R will read in the type of data as logical. In this case, the function sets the data type to be the same as the specification for the data type so the function not flag a data type error.
10.6.37.5 Value
A list having 2 components, Errors and Warnings. If no error or warning is identified, both components will contain a zero-length character string. If either an error or warning is identified, the relevant component will contain a character string that identifies the data field and the type mismatch.
10.6.38 checkModuleExists: Check whether a module required to run a model is present
10.6.38.1 Description
checkModuleExists a visioneval framework control function that checks
whether a module required to run a model is present.
10.6.38.2 Usage
checkModuleExists(ModuleName, PackageName,
InstalledPkgs_ = rownames(installed.packages()), CalledBy = NA)10.6.38.3 Arguments
| Argument | Description |
|---|---|
ModuleName |
A string identifying the module name. |
PackageName |
A string identifying the package name. |
InstalledPkgs_ |
A string vector identifying the names of packages that are installed. |
CalledBy |
A string vector having two named elements. The value of the ‘Module’ element is the name of the calling module. The value of the ‘Package’ element is the name of the package that the calling module is in. |
10.6.38.4 Details
This function takes a specified module and package, checks whether the package has been installed and whether the module is in the package. The function returns an error message is the package is not installed or if the module is not present in the package. If the module has been called by another module the value of the ‘CalledBy’ argument will be used to identify the calling module as well so that the user understands where the call is coming from.
10.6.39 checkModuleSpecs: Checks all module specifications for completeness and for incorrect entries
10.6.39.1 Description
checkModuleSpecs a visioneval framework control function that checks
all module specifications for completeness and for proper values.
10.6.39.3 Arguments
| Argument | Description |
|---|---|
Specs_ls |
a module specifications list. |
ModuleName |
a string identifying the name of the module. This is used in the error messages to identify which module has errors. |
10.6.40 checkSpec: Checks a module specifications for completeness and for incorrect entries
10.6.40.1 Description
checkSpec a visioneval framework control function that checks a single
module specification for completeness and for proper values.
10.6.40.3 Arguments
| Argument | Description |
|---|---|
Spec_ls |
a list containing the specifications for a single item in a module specifications list. |
SpecGroup |
a string identifying the specifications group the specification is in (e.g. RunBy, NewInpTable, NewSetTable, Inp, Get, Set). This is used in the error messages to identify which specification has errors. |
SpecNum |
an integer identifying which specification in the specifications group has errors. |
10.6.41 checkSpecConsistency: Check specification consistency
10.6.41.1 Description
checkSpecConsistency a visioneval framework control function that
checks whether the specifications for a dataset are consistent with the data
attributes in the datastore.
10.6.41.3 Arguments
| Argument | Description |
|---|---|
Spec_ls |
a list of data specifications consistent with a module “Get” or “Set” specifications. |
DstoreAttr_ |
a named list where the components are the attributes of a dataset. |
10.6.41.4 Details
This function compares the specifications for a dataset identified in a module “Get” or “Set” are consistent with the attributes for that data in the datastore.
10.6.41.5 Value
A list containing two components, Errors and Warnings. If no inconsistencies are found, both components will have zero-length character vectors. If there are one or more inconsistencies, then these components will hold vectors of error and warning messages. Mismatch between UNITS will produce a warning message. All other inconsistencies will produce error messages.
10.6.42 checkSpecTypeUnits: Checks the TYPE and UNITS and associated MULTIPLIER and YEAR attributes of a
Inp, Get, or Set specification for consistency.
10.6.42.1 Description
checkSpecTypeUnits a visioneval framework control function that checks
correctness of TYPE, UNITS, MULTIPLIER and YEAR attributes of a specification
that has been processed with the parseUnitsSpec function.
10.6.42.3 Arguments
| Argument | Description |
|---|---|
Spec_ls |
a list for a single specification (e.g. a Get specification for a dataset) that has been processed with the parseUnitsSpec function to split the name, multiplier, and year elements of the UNITS specification. |
SpecGroup |
a string identifying the group that this specification comes from (e.g. Inp, Get, Set). |
SpecNum |
a number identifying which specification in the order of the SpecGroup. This is used to identify the subject specification if an error is identified. |
10.6.42.4 Details
This function checks whether the TYPE and UNITS of a module’s specification contain errors. The check is done on a module specification in which the module’s UNITS attribute has been parsed by the parseUnitsSpec function to split the name, multiplier, and years parts of the UNITS attribute. The TYPE is checked against the types catalogued in the Types function. The units name in the UNITS attribute is checked against the units names corresponding to each type catalogued in the Types function. The MULTIPLIER is checked to determine whether a value is a valid number, NA, or not a number (NaN). A NA value means that no multiplier was specified (this is OK) a NaN value means that a multiplier that is not a number was specified which is an error. The YEAR attribute is checked to determine whether there is a proper specification if the specified TYPE is currency. If the TYPE is currency, a YEAR must be specified for Get and Set specifications.
10.6.43 checkTableExistence: Check whether table exists in the datastore
10.6.43.1 Description
checkTableExistence a visioneval framework control function that
checks whether a table is present in the datastore.
10.6.44 checkUnits: Check measurement units for consistency with recognized units for stated type.
10.6.44.1 Description
checkUnits a visioneval framework control function that checks the
specified UNITS for a dataset for consistency with the recognized units for
the TYPE specification for the dataset. It also splits compound units into
elements.
10.6.44.3 Arguments
| Argument | Description |
|---|---|
DataType |
a string which identifies the data type as specified in the TYPE attribute for a data set. |
Units |
a string identifying the measurement units as specified in the UNITS attribute for a data set after processing with the parseUnitsSpec function. |
10.6.44.4 Details
The visioneval code recognizes 4 simple data types (integer, double, logical, and character) and 9 complex data types (e.g. distance, time, mass). The simple data types can have any units of measure, but the complex data types must use units of measure that are declared in the Types() function. In addition, there is a compound data type that can have units that are composed of the units of two or more complex data types. For example, speed is a compound data type composed of distance divided by speed. With this example, speed in miles per hour would be represented as MI/HR. This function checks the UNITS specification for a dataset for consistency with the recognized units for the given data TYPE. To check the units of a compound data type, the function splits the units into elements and the operators that separate the elements. It identifies the element units, the complex data type for each element and the operators that separate the elements.
10.6.44.5 Value
A list which contains the following elements: DataType: a string identifying the data type. UnitType: a string identifying whether the units correspond to a ‘simple’ data type, a ‘complex’ data type, or a ‘compound’ data type. Units: a string identifying the units. Elements: a list containing the elements of a compound units. Components of this list are: Types: the complex type of each element, Units: the units of each element, Operators: the operators that separate the units. Errors: a string containing an error message or character(0) if no error.
10.6.45 convertMagnitude: Convert values between different magnitudes.
10.6.45.1 Description
convertMagnitude a visioneval framework control function that
converts values between different magnitudes such as between dollars and
thousands of dollars.
10.6.45.3 Arguments
| Argument | Description |
|---|---|
Values_ |
a numeric vector of values to convert from one unit to another. |
FromMagnitude |
a number or string identifying the magnitude of the units of the input Values_. |
ToMagnitude |
a number or string identifying the magnitude to convert the Values_ to. |
10.6.45.4 Details
The visioneval framework stores all quantities in single units to be unambiguous about the data contained in the datastore. For example, total income for a region would be stored in dollars rather than in thousands of dollars or millions of dollars. However, often inputs for large quantities are expressed in thousands or millions. Also submodels may be estimated using values expressed in multiples, or they might produce results that are multiples. Where that is the case, the framework enables model users and developers to encode the data multiplier in the input file field name or the UNITS specification. The framework functions then use that information to convert units to and from the single units stored in the datastore. This function implements the conversion. The multiplier must be specified in scientific notation used in R with the additional constraint that the digit term must be 1. For example, a multiplier of 1000 would be represented as 1e3. The multiplier is separated from the units name by a period (.). For example if the units of a dataset to be retrieved from the datastore are thousands of miles, the UNITS specification would be written as ‘MI.1e3.’
10.6.45.5 Value
A numeric vector of values corresponding the the input Values_ but converted from the magnitude identified in the FromMagnitude argument to the magnitude identified in the ToMagnitude argument. If either the FromMagnitude or the ToMagnitude arguments is NA, the original Values_ are returned. The Converted attribute of the returned values is FALSE. Otherwise the conversion is done and the Converted attribute of the returned values is TRUE.
10.6.46 convertUnits: Convert values between units of measure.
10.6.46.1 Description
convertUnits a visioneval framework control function that
converts values between different units of measure for complex and compound
data types recognized by the visioneval code.
10.6.46.3 Arguments
| Argument | Description |
|---|---|
Values_ |
a numeric vector of values to convert from one unit to another. |
DataType |
a string identifying the data type. |
FromUnits |
a string identifying the units of measure of the Values_. |
ToUnits |
a string identifying the units of measure to convert the Values_ to. If the ToUnits are ‘default’ the Values_ are converted to the default units for the model. |
10.6.46.4 Details
The visioneval code recognizes 4 simple data types (integer, double, logical, and character) and 9 complex data types (e.g. distance, time, mass). The simple data types can have any units of measure, but the complex data types must use units of measure that are declared in the Types() function. In addition, there is a compound data type that can have units that are composed of the units of two or more complex data types. For example, speed is a compound data type composed of distance divided by speed. With this example, speed in miles per hour would be represented as MI/HR. This function converts a vector of values from one unit of measure to another unit of measure. For compound data type it combines multiple unit conversions. The framework converts units based on the default units declared in the ‘units.csv’ model definition file and in UNITS specifications declared in modules.
10.6.46.5 Value
A list containing the converted values and additional information as follows: Values - a numeric vector containing the converted values. FromUnits - a string representation of the units converted from. ToUnits - a string representation of the units converted to. Errors - a string containing an error message or character(0) if no errors. Warnings - a string containing a warning message or character(0) if no warning.
10.6.47 createGeoIndex: Create datastore index.
10.6.47.1 Description
createIndex a visioneval framework control function that creates an
index for reading or writing module data to the datastore.
10.6.47.3 Arguments
| Argument | Description |
|---|---|
Table |
A string identifying the name of the table the index is being created for. |
Group |
A string identifying the name of the group where the table is located in the datastore. |
RunBy |
A string identifying the level of geography the module is being run at (e.g. Azone). |
Geo |
A string identifying the geographic unit to create the index for (e.g. the name of a particular Azone). |
GeoIndex_ls |
a list of geographic indices used to determine the positions to extract from a dataset corresponding to the specified geography. |
10.6.47.4 Details
This function creates indexing functions which return an index to positions in datasets that correspond to positions in an index field of a table. For example if the index field is ‘Azone’ in the ‘Household’ table, this function will return a function that when provided the name of a particular Azone, will return the positions corresponding to that Azone.
10.6.48 createGeoIndexList: Create a list of geographic indices for all tables in a datastore.
10.6.48.1 Description
createGeoIndexList a visioneval framework control function that
creates a list containing the geographic indices for tables in the operating
datastore for identified tables.
10.6.48.3 Arguments
| Argument | Description |
|---|---|
Specs_ls |
A ‘Get’ or ‘Set’ specifications list for a module. |
RunBy |
The value of the RunBy specification for a module. |
RunYear |
A string identifying the model year that is being run. |
10.6.48.4 Details
This function takes a ‘Get’ or ‘Set’ specifications list for a module and the ‘RunBy’ specification and returns a list which has a component for each table identified in the specifications. Each component includes all geographic datasets for the table.
10.6.49 deflateCurrency: Convert currency values to different years.
10.6.49.1 Description
deflateCurrency a visioneval framework control function that
converts currency values between different years of measure.
10.6.49.3 Arguments
| Argument | Description |
|---|---|
Values_ |
a numeric vector of values to convert from one currency year to another. |
FromYear |
a number or string identifying the currency year of the input Values_. |
ToYear |
a number or string identifying the currency year to convert the Values_ to. |
10.6.49.4 Details
The visioneval framework stores all currency values in the base year real currency (e.g. dollar) values. However, currency inputs may be in different nominal year currency. Also modules may be estimated using different nominal year currency data. For example, the original vehicle travel model in GreenSTEP used 2001 NHTS data while the newer model uses 2009 NHTS data. The framework enables model uses to specify the currency year in the field name of an input file that contains currency data. Likewise, the currency year can be encoded in the UNIT attributes for a modules Get and Set specifications. The framework converts dollars to and from specified currency year values and base year real dollar values. The model uses a set of deflator values that the user inputs for the region to make the adjustments. These values are stored in the model state list.
10.6.49.5 Value
A numeric vector of values corresponding the the input Values_ but converted from the currency year identified in the FromYear argument to the currency year identified in the ToYear argument. If either the FromYear or the ToYear arguments is unaccounted for in the deflator series, the original Values_ are returned with a Converted attribute of FALSE. Otherwise the conversion is done and the Converted attribute of the returned values is TRUE.
10.6.50 doProcessInpSpec: Filters Inp specifications list based on OPTIONAL specification attributes.
10.6.50.1 Description
doProcessInpSpec a visioneval framework control function that filters
out Inp specifications whose OPTIONAL specification attribute is TRUE but the
specified input file is not present.
10.6.50.3 Arguments
| Argument | Description |
|---|---|
InpSpecs_ls |
A standard specifications list for Inp specifications. |
InputDir |
The path to the input directory. |
10.6.50.4 Details
An Inp specification component may have an OPTIONAL specification whose value is TRUE. If so, and if the specified input file is present, then the input specification needs to be processed. This function checks whether the OPTIONAL specification is present, whether its value is TRUE, and whether the file exists. If all of these are true, then the input specification needs to be processed. The input specification also needs to be processed if it is not optional. A specification is not optional if the OPTIONAL attribute is not present or if it is present and the value is not TRUE. The function returns a list of all the Inp specifications that meet these criteria.
10.6.51 expandSpec: Expand a Inp, Get, or Set specification so that is can be used by other
functions to process inputs and to read from or write to the datastore.
10.6.51.1 Description
expandSpec a visioneval framework control function that takes a Inp,
Get, or Set specification and processes it to be in a form that can be used
by other functions which use the specification in processing inputs or
reading from or writing to the datastore. The parseUnitsSpec function is
called to parse the UNITS attribute to extract name, multiplier, and year
values. When the specification has multiple values for the NAME attribute,
the function creates a specification for each name value.
10.6.51.3 Arguments
| Argument | Description |
|---|---|
SpecToExpand_ls |
A standard specifications list for a specification whose NAME attribute has multiple values. |
ComponentName |
A string that is the name of the specifications that the specification is a part of (e.g. “Inp,” “Get,” “Set”). |
10.6.51.4 Details
The VisionEval design allows module developers to assign multiple values to the NAME attributes of a Inp, Get, or Set specification where the other attributes for those named datasets (or fields) are the same. This greatly reduces duplication and the potential for error in writing module specifications. However, other functions that check or use the specifications are not capable of handling specifications which have NAME attributes containing multiple values. This function expands a specification with multiple values for a NAME attribute into multiple specifications, each with a single value for the NAME attribute. In addition, the function calls the parseUnitsSpec function to extract multiplier and year information from the value of the UNITS attribute. See that function for details.
10.6.52 findSpec: Find the full specification corresponding to a defined NAME, TABLE, and GROUP
10.6.52.1 Description
findSpec a visioneval framework control function that returns the full
dataset specification for defined NAME, TABLE, and GROUP.
10.6.52.3 Arguments
| Argument | Description |
|---|---|
Specs_ls |
a standard specifications list for ‘Inp,’ ‘Get,’ or ‘Set’ |
Name |
a string for the name of the dataset |
Table |
a string for the table that the dataset resides in |
Group |
a string for the generic group that the table resides in |
10.6.52.4 Details
This function finds and returns the full specification from a specifications list whose NAME, TABLE and GROUP values correspond to the Name, Table, and Group argument values. The specifications list must be in standard format and must be for only ‘Inp,’ ‘Get,’ or ‘Set’ specifications.
10.6.53 getDatasetAttr: Get attributes of a dataset
10.6.53.1 Description
getDatasetAttr a visioneval framework control function that retrieves
the attributes for a dataset in the datastore.
10.6.53.3 Arguments
| Argument | Description |
|---|---|
Name |
a string identifying the dataset name. |
Table |
a string identifying the table the dataset is a part of. |
Group |
a string or numeric representation of the group the table is a part of. |
DstoreListing_df |
a dataframe which lists the contents of the datastore as contained in the model state file. |
10.6.54 getFromDatastore: Retrieve data identified in ‘Get’ specifications from datastore
10.6.54.1 Description
getFromDatastore a visioneval framework control function that
retrieves datasets identified in a module’s ‘Get’ specifications from the
datastore.
10.6.54.3 Arguments
| Argument | Description |
|---|---|
ModuleSpec_ls |
a list of module specifications that is consistent with the VisionEval requirements |
RunYear |
a string identifying the model year being run. The default is the Year object in the global workspace. |
Geo |
a string identifying the name of the geographic area to get the data for. For example, if the module is specified to be run by Azone, then Geo would be the name of a particular Azone. |
GeoIndex_ls |
a list of geographic indices used to determine the positions to extract from a dataset corresponding to the specified geography. |
10.6.54.4 Details
This function retrieves from the datastore all of the data sets identified in a module’s ‘Get’ specifications. If the module’s specifications include the name of a geographic area, then the function will retrieve the data for that geographic area.
10.6.55 getModelState: Get values from model state list.
10.6.55.1 Description
getModelState a visioneval framework control function that reads
components of the list that keeps track of the model state.
10.6.55.3 Arguments
| Argument | Description |
|---|---|
Names_ |
A string vector of the components to extract from the ModelState_ls list. |
10.6.56 getModuleSpecs: Retrieve module specifications from a package
10.6.56.1 Description
getModuleSpecs a visioneval framework control function that retrieves
the specifications list for a module and returns the specifications list.
10.6.56.3 Arguments
| Argument | Description |
|---|---|
ModuleName |
A string identifying the name of the module. |
PackageName |
A string identifying the name of the package that the module is in. |
10.6.56.4 Details
This function loads the specifications for a module in a package. It returns the specifications list.
10.6.57 getUnits: Retrieve default units for model
10.6.57.1 Description
getUnits a visioneval framework control function that retrieves the
default model units for a vector of complex data types.
10.6.57.3 Arguments
| Argument | Description |
|---|---|
Type_ |
A string vector identifying the complex data type(s). |
10.6.57.4 Details
This is a convenience function to make it easier to retrieve the default units for a complex data type (e.g. distance, volume, speed). The default units are the units used to store the complex data type in the datastore.
10.6.58 initDatastoreGeography: Initialize datastore geography.
10.6.58.1 Description
initDatastoreGeography a visioneval framework control function that
initializes tables and writes datasets to the datastore which describe
geographic relationships of the model.
10.6.58.3 Details
This function writes tables to the datastore for each of the geographic levels. These tables are then used during a model run to store values that are either specified in scenario inputs or that are calculated during a model run. The function populates the tables with cross-references between geographic levels. The function reads the model geography (Geo_df) from the model state file. Upon successful completion, the function calls the listDatastore function to update the datastore listing in the global list.
10.6.59 initLog: Initialize run log.
10.6.59.1 Description
initLog a visioneval framework control function that creates a log
(text file) that stores messages generated during a model run.
10.6.59.3 Arguments
| Argument | Description |
|---|---|
Suffix |
A character string appended to the file name for the log file. For example, if the suffix is ‘CreateHouseholds,’ the log file is named ‘Log_CreateHouseholds.txt.’ The default value is NULL in which case the suffix is the date and time. |
10.6.59.4 Details
This function creates a log file that is a text file which stores messages
generated during a model run. The name of the log is ‘Log
10.6.60 initModelStateFile: Initialize model state.
10.6.60.1 Description
initModelState a visioneval framework control function that loads
model run parameters into the model state list in the global workspace and
saves as file.
10.6.60.2 Usage
initModelStateFile(Dir = "defs", ParamFile = "run_parameters.json",
DeflatorFile = "deflators.csv", UnitsFile = "units.csv")10.6.60.3 Arguments
| Argument | Description |
|---|---|
Dir |
A string identifying the name of the directory where the global parameters, deflator, and default units files are located. The default value is defs. |
ParamFile |
A string identifying the name of the global parameters file. The default value is parameters.json. |
DeflatorFile |
A string identifying the name of the file which contains deflator values by year (e.g. consumer price index). The default value is deflators.csv. |
UnitsFile |
A string identifying the name of the file which contains default units for complex data types (e.g. currency, distance, speed, etc.). The default value is units.csv. |
10.6.60.4 Details
This function creates the model state list and loads model run parameters recorded in the ‘parameters.json’ file into the model state list. It also saves the model state list in a file (ModelState.Rda).
10.6.61 inputsToDatastore: Write the datasets in a list of module inputs that have been processed to the
datastore.
10.6.61.1 Description
inputsToDatastore a visioneval framework control function that takes a
list of processed module input files and writes the datasets to the
datastore.
10.6.61.3 Arguments
| Argument | Description |
|---|---|
Inputs_ls |
a list processes module inputs as created by the ‘processModuleInputs’ function. |
ModuleSpec_ls |
a list of module specifications that is consistent with the VisionEval requirements. |
ModuleName |
a string identifying the name of the module (used to document the dataset in the datastore). |
10.6.61.4 Details
This function takes a processed list of input datasets specified by a module created by the application of the ‘processModuleInputs’ function and writes the datasets in the list to the datastore.
10.6.62 loadDatastore: Load saved datastore.
10.6.62.1 Description
loadDatastore a visioneval framework control function that copies an
existing saved datastore and writes information to run environment.
10.6.62.3 Arguments
| Argument | Description |
|---|---|
FileToLoad |
A string identifying the full path name to the saved datastore. Path name can either be relative to the working directory or absolute. |
Dir |
A string identifying the path of the geography definition file (GeoFile), default to ‘defs’ relative to the working directory |
GeoFile |
A string identifying the name of the geography definition file (see ‘readGeography’ function) that is consistent with the saved datastore. The geography definition file must be located in the ‘defs’ directory. |
SaveDatastore |
A logical identifying whether an existing datastore will be saved. It is renamed by appending the system time to the name. The default value is TRUE. |
10.6.62.4 Details
This function copies a saved datastore as the working datastore attributes the global list with related geographic information. This function enables scenario variants to be built from a constant set of starting conditions.
10.6.62.5 Value
TRUE if the datastore is loaded. It copies the saved datastore to working directory as ‘datastore.h5.’ If a ‘datastore.h5’ file already exists, it first renames that file as ‘archive-datastore.h5.’ The function updates information in the model state file regarding the model geography and the contents of the loaded datastore. If the stored file does not exist an error is thrown.
10.6.63 loadModelParameters: Load model global parameters file into datastore.
10.6.63.1 Description
loadModelParameters a visioneval framework control function reads the
‘model_parameters.json’ file and stores the contents in the ‘Global/Model’
group of the datastore.
10.6.63.3 Arguments
| Argument | Description |
|---|---|
ModelParamFile |
A string identifying the name of the parameter file. The default value is ‘model_parameters.json.’ |
10.6.63.4 Details
This function reads the ‘model_parameters.json’ file in the ‘defs’ directory which contains parameters specific to a model rather than to a module. These area parameters that may be used by any module. Parameters are specified by name, value, and data type. The function creates a ‘Model’ group in the ‘Global’ group and stores the values of the appropriate type in the ‘Model’ group.
10.6.64 parseInputFieldNames: Parse field names of input file to separate out the field name, currency
year, and multiplier.
10.6.64.1 Description
parseInputFieldNames a visioneval framework control function that
parses the field names of an input file to separate out the field name,
currency year (if data is currency type), and value multiplier.
10.6.64.3 Arguments
| Argument | Description |
|---|---|
FieldNames_ |
A character vector containing the field names of an input file. |
Specs_ls |
A list of specifications for fields in the input file. |
FileName |
A string identifying the name of the file that the field names are from. This is used for writing error messages. |
10.6.64.4 Details
The field names of input files can be used to encode more information than the name itself. It can also encode the currency year for currency type data and also if the values are in multiples (e.g. thousands of dollars). For currency type data it is mandatory that the currency year be specified so that the data can be converted to base year currency values (e.g. dollars in base year dollars). The multiplier is optional, but needless to say, it can only be applied to numeric data. The function returns a list with a component for each field. Each component identifies the field name, year, multiplier, and error status for the result of parsing the field name. If the field name was parsed successfully, the error status is character(0). If the field name was not successfully parsed, the error status contains an error message, identifying the problem.
10.6.64.5 Value
A named list with one component for each field. Each component is a list having 4 named components: Error, Name, Year, Multiplier. The Error component has a value of character(0) if there are no errors or a character vector of error messages if there are errors. The Name component is a string with the name of the field. The Year component is a string with the year component if the data type is currency or NA if the data type is not currency or if the Year component has an invalid value. The Multiplier is a number if the multiplier component is present and is valid. It is NA if there is no multiplier component and NaN if the multiplier is invalid. Each component of the list is named with the value of the Name component (i.e. the field name without the year and multiplier elements.)
10.6.65 parseModelScript: Parse model script.
10.6.65.1 Description
parseModel a visioneval framework control function that reads and
parses the model script to identify the sequence of module calls and the
associated call arguments.
10.6.65.3 Arguments
| Argument | Description |
|---|---|
FilePath |
A string identifying the relative or absolute path to the model run script is located. |
TestMode |
A logical identifying whether the function is to run in test mode. When in test mode the function returns the parsed script but does not change the model state or write results to the log. |
10.6.65.4 Details
This function reads in the model run script and parses the script to identify the sequence of module calls. It extracts each call to ‘runModule’ and identifies the values assigned to the function arguments. It creates a list of the calls with their arguments in the order of the calls in the script.
10.6.66 parseUnitsSpec: Parse units specification into components and add to specifications list.
10.6.66.1 Description
parseUnitsSpec a visioneval framework control function that parses the
UNITS attribute of a standard Inp, Get, or Set specification for a dataset to
identify the units name, multiplier, and year for currency data. Returns a
modified specifications list whose UNITS value is only the units name, and
includes a MULTIPLIER attribute and YEAR attribute.
10.6.66.3 Arguments
| Argument | Description |
|---|---|
Spec_ls |
A standard specifications list for a Inp, Get, or Set item. |
ComponentName |
A string that is the name of the specifications the the specification comes from (e.g. “Inp,” “Get,” “Set). |
10.6.66.4 Details
The UNITS component of a specifications list can encode information in addition to the units name. This includes a value units multiplier and in the case of currency values the year for the currency measurement. The multiplier element can only be expressed in scientific notation where the number before the ‘e’ can only be 1. If the year element for a currency specification is missing, it is replaced by the model base year which is recorded in the model state file. If this is done, a WARN attribute is added to the specifications list notifying the module developer that there is no year element and the model base year will be used when the module is called. The test module function reads this warning and writes it to the module test log. This way the module developer is made aware of the situation so that it may be corrected if necessary. The model user is not bothered by the warning.
10.6.66.5 Value
a list that is a standard specifications list with the addition of a MULTIPLIER component and a YEAR component as well as a modification of the UNIT component. The MULTIPLIER component can have the value of NA, a number, or NaN. The value is NA if the multiplier is missing. It is a number if the multiplier is a valid number. The value is NaN if the multiplier is not a valid number. The YEAR component is a character string that is a 4-digit representation of a year or NA if the component is not a proper year. If the year component is missing from the UNITS specification for currency data, the model base year is substituted. In that case, a WARN attribute is added to the returned specifications list. This is read by the testModule function and written to the module test log to notify the module developer. After the UNITS component has been parsed and the YEAR and MULTIPLIER components extracted, the UNITS component is modified to only be the units name.
10.6.67 processModuleInputs: Process module input files
10.6.67.1 Description
processModuleInputs a visioneval framework control function that
processes input files identified in a module’s ‘Inp’ specifications in
preparation for saving in the datastore.
10.6.67.3 Arguments
| Argument | Description |
|---|---|
ModuleSpec_ls |
a list of module specifications that is consistent with the VisionEval requirements. |
ModuleName |
a string identifying the name of the module (used to document module in error messages). |
Dir |
a string identifying the relative path to the directory where the model inputs are contained. |
10.6.67.4 Details
This function processes the input files identified in a module’s ‘Inp’ specifications in preparation for saving the data in the datastore. Several processes are carried out. The existence of each specified input file is checked. Any file whose corresponding ‘GROUP’ specification is ‘Year,’ is checked to determine that it has ‘Year’ and ‘Geo’ columns. The entries in the ‘Year’ and ‘Geo’ columns are checked to make sure they are complete and there are no duplicates. Any file whose ‘GROUP’ specification is ‘Global’ or ‘BaseYear’ and whose ‘TABLE’ specification is a geographic specification other than ‘Region’ is checked to determine if it has a ‘Geo’ column and the entries are checked for completeness. The data in each column are checked against specifications to determine conformance. The function returns a list which contains a list of error messages and a list of the data inputs. The function also writes error messages and warnings to the log file.
10.6.67.5 Value
A list containing the results of the input processing. The list has two components. The first (Errors) is a vector of identified file and data errors. The second (Data) is a list containing the data in the input files organized in the standard format for data exchange with the datastore.
10.6.68 processModuleSpecs: Process module specifications to expand items with multiple names.
10.6.68.1 Description
processModuleSpecs a visioneval framework control function that
processes a full module specifications list, expanding all elements in the
Inp, Get, and Set components by parsing the UNITS attributes and duplicating
every specification which has multiple values for the NAME attribute.
10.6.69 readGeography: Read geographic specifications.
10.6.69.1 Description
readGeography a visioneval framework control function that reads the
geographic specifications file for the model.
10.6.69.3 Arguments
| Argument | Description |
|---|---|
Dir |
A string identifying the path to the geographic specifications file. Note: don’t include the final separator in the path name ‘e.g. not defs/.’ |
GeoFile |
A string identifying the name of the geographic specifications file. This is a csv-formatted text file which contains columns named ‘Azone,’ ‘Bzone,’ ‘Czone,’ and ‘Marea.’ The ‘Azone’ column must have zone names in all rows. The ‘Bzone’ and ‘Czone’ columns can be unspecified (NA in all rows) or may have have unique names in every row. The ‘Marea’ column (referring to metropolitan areas) identifies metropolitan areas corresponding to the most detailed level of specified geography (or ‘None’ no metropolitan area occupies any portion of the zone. |
10.6.69.4 Details
This function manages the reading and error checking of geographic specifications for the model. It calls the checkGeography function to check for errors in the specifications. The checkGeography function reads in the file and checks for errors. It returns a list of any errors that are found and a data frame containing the geographic specifications. If errors are found, the functions writes the errors to a log file and stops model execution. If there are no errors, the function adds the geographic in the geographic specifications file, the errors are written to the log file and execution stops. If no errors are found, the geographic specifications are added to the model state file.
10.6.69.5 Value
The value TRUE is returned if the function is successful at reading the file and the specifications are consistent. It stops if there are any errors in the specifications. All of the identified errors are written to the run log. A data frame containing the file entries is added to the model state file as Geo_df’.
10.6.70 readModelState: Reads values from model state file.
10.6.70.1 Description
readModelState a visioneval framework control function that reads
components of the file that saves a copy of the model state.
10.6.70.3 Arguments
| Argument | Description |
|---|---|
Names_ |
A string vector of the components to extract from the ModelState_ls list. |
FileName |
A string vector with the full path name of the model state file. |
10.6.71 setInDatastore: Save the data sets returned by a module in the datastore
10.6.71.1 Description
setInDatastore a visioneval framework control function saves to the
datastore the data returned in a standard list by a module.
10.6.71.2 Usage
setInDatastore(Data_ls, ModuleSpec_ls, ModuleName, Year, Geo = NULL,
GeoIndex_ls = NULL)10.6.71.3 Arguments
| Argument | Description |
|---|---|
Data_ls |
a list containing the data to be saved. The list is organized by group, table, and data set. |
ModuleSpec_ls |
a list of module specifications that is consistent with the VisionEval requirements |
ModuleName |
a string identifying the name of the module (used to document the module creating the data in the datastore) |
Year |
a string identifying the model run year |
Geo |
a string identifying the name of the geographic area to get the data for. For example, if the module is specified to be run by Azone, then Geo would be the name of a particular Azone. |
GeoIndex_ls |
a list of geographic indices used to determine the positions to extract from a dataset corresponding to the specified geography. |
10.6.71.4 Details
This function saves to the datastore the data sets identified in a module’s ‘Set’ specifications and included in the list returned by the module. If a particular geographic area is identified, the data are saved to the positions in the data sets in the datastore corresponding to the identified geographic area.
10.6.72 setModelState: Update model state.
10.6.72.1 Description
setModelState a visioneval framework control function that updates the
list that keeps track of the model state with list of components to update
and resaves in the model state file.
10.6.72.3 Arguments
| Argument | Description |
|---|---|
ChangeState_ls |
A named list of components to change in ModelState_ls |
FileName |
A string identifying the name of the file that contains the ModelState_ls list. The default name is ‘ModelState.Rda.’ |
10.6.72.4 Details
Key variables that are important for managing the model run are stored in a list (ModelState_ls) that is in the global workspace and saved in the ‘ModelState.Rda’ file. This function updates entries in the model state list with a supplied named list of values, and then saves the results in the file.
10.6.73 simDataTransactions: Create simulation of datastore transactions.
10.6.73.1 Description
simDataTransactions a visioneval framework control function that loads
all module specifications in order (by run year) and creates a simulated
listing of the data which is in the datastore and the requests of data from
the datastore and checks whether tables will be present to put datasets in
and that datasets will be present that data is to be retrieved from.
10.6.73.3 Arguments
| Argument | Description |
|---|---|
AllSpecs_ls |
A list containing the processed specifications of all of the modules run by model script in the order that the modules are called with duplicated module calls removed. Information about each module call is a component of the list in the order of the module calls. Each component is composed of 3 components: ‘ModuleName’ contains the name of the module, ‘PackageName’ contains the name of the package the module is in, and ‘Specs’ contains the processed specifications of the module. The ‘Get’ specification component includes the ‘Get’ specifications of all modules that are called by the module. |
10.6.73.4 Details
This function creates a list of the datastore listings for the working datastore and for all datastore references. The list includes a ‘Global’ component, in which ‘Global’ references are simulated, components for each model run year, in which ‘Year’ references are simulated, and if the base year is not one of the run years, a base year component, in which base year references are simulated. For each model run year the function steps through a data frame of module calls as produced by ‘parseModelScript,’ and loads and processes the module specifications in order: adds ‘NewInpTable’ references, adds ‘Inp’ dataset references, checks whether references to datasets identified in ‘Get’ specifications are present, adds ‘NewSetTable’ references, and adds ‘Set’ dataset references. The function compiles a vector of error and warning messages. Error messages are made if: 1) a ‘NewInpTable’ or ‘NewSetTable’ specification of a module would create a new table for a table that already exists; 2) a dataset identified by a ‘Get’ specification would not be present in the working datastore or any referenced datastores; 3) the ‘Get’ specifications for a dataset would not be consistent with the specifications for the dataset in the datastore. The function compiles warnings if a ‘Set’ specification will cause existing data in the working datastore to be overwritten. The function writes warning and error messages to the log and stops program execution if there are any errors.
10.6.74 sortGeoTable: Sort a data frame so that the order of rows matches the geography in a
datastore table.
10.6.74.1 Description
sortGeoTable a visioneval framework control function that returns a
data frame whose rows are sorted to match the geography in a specified table
in the datastore.
10.6.74.3 Arguments
| Argument | Description |
|---|---|
Data_df |
a data frame that contains a ‘Geo’ field containing the names of the geographic areas to sort by and any number of additional data fields. |
Table |
a string for the table that is to be matched against. |
Group |
a string for the generic group that the table resides in. |
10.6.74.4 Details
This function sorts the rows of a data frame that the ‘Geo’ field in the data frame matches the corresponding geography names in the specified table in the datastore. The function returns the sorted table.
10.6.75 SpecRequirements: List basic module specifications to check for correctness
10.6.75.1 Description
SpecRequirements a visioneval framework control function that returns
a list of basic requirements for module specifications to be used for
checking correctness of specifications.
10.6.75.3 Details
This function returns a list of the basic requirements for module specifications. The main components of the list are the components of module specifications: RunBy, NewInpTable, NewSetTable, Inp, Get, Set. For each item of each module specifications component, the list identifies the required data type of the attribute entry and the allowed values for the attribute entry.
10.6.75.4 Value
A list comprised of six named components: RunBy, NewInpTable, NewSetTable, Inp, Get, Set. Each main component is a list that has a component for each specification item that has values to be checked. For each such item there is a list having two components: ValueType and ValuesAllowed. The ValueType component identifies the data type that the data entry for the item must have (e.g. character, integer). The ValuesAllowed item identifies what values the item may have.
10.6.76 Types: Returns a list of returns a list of recognized data types, the units for each
type, and storage mode of each type.
10.6.76.1 Description
Types a visioneval framework control function that returns a list of
returns a list of recognized data types, the units for each type, and storage
mode of each type.
10.6.76.3 Details
This function stores a listing of the dataset types recognized by the visioneval framework, the units recognized for each type, and the storage mode used for each type. Types include simple types (e.g. integer, double, character, logical) as well as complex types (e.g. distance, time, mass). For the complex types, units are specified as well. For example for the distance type, allowed units are MI (miles), FT (feet), KM (kilometers), M (meters). The listing includes conversion factors between units of each complex type. The listing also contains the storage mode (i.e. integer, double, character, logical of each type. For simple types, the type and the storage mode are the same).
10.6.76.4 Value
A list containing a component for each recognized type. Each component lists the recognized units for the type and the storage mode. There are currently 4 simple types and 10 complex type. The simple types are integer, double, character and logical. The complex types are currency, distance, area, mass, volume, time, speed, vehicle_distance, passenger_distance, and payload_distance.
10.6.77 writeLog: Write to log.
10.6.77.1 Description
writeLog a visioneval framework control function that writes a message
to the run log.
10.6.77.3 Arguments
| Argument | Description |
|---|---|
Msg |
A character string. |
Print |
logical (default: FALSE). If True Msg will be printed in additon to being added to log |
10.6.77.4 Details
This function writes a message in the form of a string to the run log. It logs the time as well as the message to the run log.
10.6.78 writeVENameRegistry: Writes module Inp and Set specifications to the VisionEval name registry.
10.6.78.1 Description
writeVENameRegistry a visioneval framework control function that
writes module Inp and Set specifications to the VisionEval name registry.
10.6.78.3 Arguments
| Argument | Description |
|---|---|
ModuleName |
a string identifying the module name. |
PackageName |
a string identifying the package name. |
NameRegistryDir |
a string identifying the path to the directory where the name registry file is located. |
10.6.78.4 Details
The VisionEval name registry (VENameRegistry.json) keeps track of the dataset names created by all registered modules by reading in datasets specified in the module Inp specifications or by returning calculated datasets as specified in the module Set specifications. This functions adds the Inp and Set specifications for a module to the registry. It removes any existing entries for the module first.
10.6.80 initDatasetH5: Initialize dataset in an HDF5 (H5) type datastore table.
10.6.80.1 Description
initDatasetH5 a visioneval framework datastore connection function
that initializes a dataset in an HDF5 (H5) type datastore table.
10.6.80.3 Arguments
| Argument | Description |
|---|---|
Spec_ls |
a list containing the standard module ‘Set’ specifications described in the model system design documentation. |
Group |
a string representation of the name of the group the table is to be created in. |
10.6.81 initDatasetRD: Initialize dataset in an RData (RD) type datastore table.
10.6.81.1 Description
initDatasetRD a visioneval framework datastore connection function
initializes a dataset in an RData (RD) type datastore table.
10.6.81.3 Arguments
| Argument | Description |
|---|---|
Spec_ls |
a list containing the standard module specifications described in the model system design documentation. |
Group |
a string representation of the name of the top-level subdirectory the table is to be created in (i.e. either ‘Global’ or the name of the year). |
10.6.82 initDatastoreH5: Initialize Datastore for an HDF5 (H5) type datastore.
10.6.82.1 Description
initDatastoreH5 a visioneval framework datastore connection function
that creates datastore with starting structure for an HDF5 (H5) type
datastore.
10.6.82.3 Details
This function creates the datastore for the model run with the initial structure for an HDF5 (H5) type datastore.
10.6.83 initDatastoreRD: Initialize Datastore for an RData (RD) type datastore.
10.6.83.1 Description
initDatastoreRD a visioneval framework datastore connection function
that creates a datastore with starting structure for an RData (RD) type
datastore.
10.6.83.3 Details
This function creates the datastore for the model run with the initial structure for an RData (RD) type datastore.
10.6.84 initTableH5: Initialize table in an HDF5 (H5) type datastore.
10.6.84.1 Description
initTableH5 a visioneval framework datastore connection function that
initializes a table in an HDF5 (H5) type datastore.
10.6.84.3 Arguments
| Argument | Description |
|---|---|
Table |
a string identifying the name of the table to initialize. |
Group |
a string representation of the name of the group the table is to be created in. |
Length |
a number identifying the table length. |
10.6.84.5 Value
The value TRUE is returned if the function is successful at creating the table. In addition, the listDatastore function is run to update the inventory in the model state file. The function stops if the group in which the table is to be placed does not exist in the datastore and a message is written to the log.
10.6.85 initTableRD: Initialize table in an RData (RD) type datastore.
10.6.85.1 Description
initTableRD a visioneval framework datastore connection function
initializes a table in an RData (RD) type datastore.
10.6.85.3 Arguments
| Argument | Description |
|---|---|
Table |
a string identifying the name of the table to initialize. |
Group |
a string representation of the name of the top-level subdirectory the table is to be created in (i.e. either ‘Global’ or the name of the year). |
Length |
a number identifying the table length. |
10.6.85.5 Value
The value TRUE is returned if the function is successful at creating the table. In addition, the listDatastore function is run to update the inventory in the model state file. The function stops if the group in which the table is to be placed does not exist in the datastore and a message is written to the log.
10.6.86 listDatastoreH5: List datastore contents for an HDF5 (H5) type datastore.
10.6.86.1 Description
listDatastoreH5 a visioneval framework datastore connection function
that lists the contents of an HDF5 (H5) type datastore.
10.6.87 listDatastoreRD: List datastore contents for an RData (RD) type datastore.
10.6.87.1 Description
listDatastoreRD a visioneval framework datastore connection function
that lists the contents of an RData (RD) type datastore.
10.6.87.3 Arguments
| Argument | Description |
|---|---|
DataListing_ls |
a list containing named elements describing a new data item being added to the datastore listing and the model state file. The list components are: group - the name of the group (path) the item is being added to; name - the name of the data item (directory or dataset); groupname - the full path to the data item; attributes - a list containing the named attributes of the data item. |
10.6.88 readFromTableH5: Read from an HDF5 (H5) type datastore table.
10.6.88.1 Description
readFromTableH5 a visioneval framework datastore connection function
that reads a dataset from an HDF5 (H5) type datastore table.
10.6.88.3 Arguments
| Argument | Description |
|---|---|
Name |
A string identifying the name of the dataset to be read from. |
Table |
A string identifying the complete name of the table where the dataset is located. |
Group |
a string representation of the name of the datastore group the data is to be read from. |
File |
a string representation of the file path of the datastore |
Index |
A numeric vector identifying the positions the data is to be written to. NULL if the entire dataset is to be read. |
ReadAttr |
A logical identifying whether to return the attributes of the stored dataset. The default value is FALSE. |
10.6.89 readFromTableRD: Read from an RData (RD) type datastore table.
10.6.89.1 Description
readFromTableRD a visioneval framework datastore connection function
that reads a dataset from an RData (RD) type datastore table.
10.6.89.2 Usage
readFromTableRD(Name, Table, Group, DstoreLoc = NULL, Index = NULL,
ReadAttr = FALSE)10.6.89.3 Arguments
| Argument | Description |
|---|---|
Name |
A string identifying the name of the dataset to be read from. |
Table |
A string identifying the complete name of the table where the dataset is located. |
Group |
a string representation of the name of the datastore group the data is to be read from. |
DstoreLoc |
a string representation of the file path of the datastore. NULL if the datastore is the current directory. |
Index |
A numeric vector identifying the positions the data is to be written to. NULL if the entire dataset is to be read. |
ReadAttr |
A logical identifying whether to return the attributes of the stored dataset. The default value is FALSE. |
10.6.90 writeToTableH5: Write to an RData (RD) type datastore table.
10.6.90.1 Description
writeToTableRD a visioneval framework datastore connection function
that writes data to an RData (RD) type datastore table and initializes
dataset if needed.
10.6.90.3 Arguments
| Argument | Description |
|---|---|
Data_ |
A vector of data to be written. |
Spec_ls |
a list containing the standard module ‘Set’ specifications described in the model system design documentation. |
Group |
a string representation of the name of the datastore group the data is to be written to. |
Index |
A numeric vector identifying the positions the data is to be written to. |
10.6.91 writeToTableRD: Write to an RData (RD) type datastore table.
10.6.91.1 Description
writeToTableRD a visioneval framework datastore connection function
that writes data to an RData (RD) type datastore table and initializes
dataset if needed.
10.6.91.3 Arguments
| Argument | Description |
|---|---|
Data_ |
A vector of data to be written. |
Spec_ls |
a list containing the standard module ‘Set’ specifications described in the model system design documentation. |
Group |
a string representation of the name of the datastore group the data is to be written to. |
Index |
A numeric vector identifying the positions the data is to be written to. |