Setting linear models

.gitignore

 .buildpath
 .project
+*.so
 

README.md

@@ -9,15 +9,16 @@ Supported:
 
 For simple usage see tests:
 
-1. Install requirements (run once):
-	`$ julia scripts/requirements.jl`
+1. Install requirements and precompile dependencies (run once). It takes time the first time, 
+   generates `jexio_deps.so`:
+    - `$ julia scripts/requirements.jl`
 
 2. Run tests:
-	`$ julia --project=. test/runtests.jl`
-   Optionally, when developing run tests from the Julia REPL for preloading libraries, 
+    - `$ julia -Jjexio_deps.so --project=. test/runtests.jl`
+   Optionally, run tests from the Julia REPL for preloading libraries, 
    as it results in faster runtimes after the first time.
-    `$ julia --project=.`
-    `julia> include("test/runtests.jl")
+    - `$ julia -Jjexio_deps.so --project=.`
+    - `julia> include("test/runtests.jl")
 
 Formatting the code:
 

scripts/requirements.jl

@@ -10,5 +10,10 @@ Pkg.add("Glob")
 Pkg.add("DataFrames")
 Pkg.add("GLM")
 Pkg.add("Plots")
+Pkg.add("PackageCompiler")
+
+
+using PackageCompiler
+create_sysimage([:Glob, :Plots, :GLM, :DataFrames], sysimage_path = "jexio_deps.so")
 
 exit()

src/Exio.jl

 module Exio
 
-export ExioH, exio_init, input_parser
+export ExioH, exio_init
 
 include("extractor/Extractor.jl")
 
 mutable struct ExioH
     extractor::AbstractExtractor
-
     ExioH() = new()
 end
 
@@ -16,8 +15,7 @@ function exio_init(app::String, outputPrefix::String)::ExioH
     exioH = ExioH()
     if app == "AmrexCastro"
         println("Hello AmrexCastro")
-        exioH.extractor = AmrexCastro()
-        init!(exioH.extractor, outputPrefix)
+        exioH.extractor = AmrexCastro(outputPrefix)
     end
 
     return exioH

src/extractor/Amrex.jl

@@ -4,23 +4,17 @@ abstract type AbstractAmrex <: AbstractExtractor end
 mutable struct Amrex <: AbstractAmrex
     app::String
     outputPrefix::String
-    degreesOfFreedom::Any
-
-    Amrex() = new()
+    inputs::Array{String}
+    outputs::Array{String}
+
+    Amrex(outputPrefix::String) = new(
+        "Amrex",
+        outputPrefix,
+        ["max_step", "amr.check_int", "amr.plot_int", "amr.n_cell", "amr.max_level"],
+        ["plots_size", "checkpoints_size"],
+    )
 end
 
-"""
-   
-   initialize members of the extractor::Amrex type
-   Using bang convention as init modifies the extractor::Amrex
-   https://docs.julialang.org/en/v1/manual/style-guide/index.html#bang-convention-1
-"""
-function init!(extractor::Amrex, outputPrefix::String)
-    extractor.app = "Amrex"
-    extractor.outputPrefix = outputPrefix
-    extractor.degreesOfFreedom =
-        ["max_step", "amr.check_int", "amr.plot_int", "amr.n_cell", "amr.max_level"]
-end
 
 """
     Parses an input file with entries key = value, returns all entries in a Dict
@@ -28,7 +22,7 @@ end
     - `extractor::AbstractAmrex` : input type extending AbstractAmrex
     - `inputFile::String` : input file to be parsed, absolute path is preferred
 """
-function input_parser(extractor::AbstractAmrex, inputFile::String)::Dict{String,String}
+function _input_parser(extractor::AbstractAmrex, inputFile::String)::Dict{String,String}
 
     parameters = Dict{String,String}()
     # get file contents in a single iterable type, \n newline is removed 
@@ -64,24 +58,25 @@ end
     from input parameters 
     # Arguments
     - `extractor::AbstractAmrex` : input type extending AbstractAmrex
-    - `parameters::Dict{String,String}` : input file to be parsed, absolute path is preferred
+    - `parameters::Dict{Any,Any}` : input file to be parsed, absolute path is preferred
 """
-function get_independent_variables(extractor::AbstractAmrex, parameters)::Dict
+function _get_independent_variables(extractor::AbstractAmrex, parameters)::Dict{String,Any}
 
-    independentVariables = Dict()
+    independentVariables = Dict{String,Any}()
 
     # Number of output events, 
     # get is the safe way. 3rd argument is a default fallback if key not found
     maxStepStr = get(parameters, "max_step", "")
 
     if maxStepStr != ""
-        
-      # safe string conversion to a type
+
+        # safe string conversion to a type
         maxStep = parse(Int64, maxStepStr)
 
         plotIntStr = get(parameters, "amr.plot_int", "")
         if plotIntStr != ""
             plotFrequency = parse(Int64, plotIntStr)
+            independentVariables["amr.plot_int"] = plotFrequency
             independentVariables["amr.nplot_files"] = floor(Int32, maxStep / plotFrequency)
         else
             throw(NoSuchFieldException("jexio Amrex: Can't find amr.plot_int"))
@@ -90,7 +85,9 @@ function get_independent_variables(extractor::AbstractAmrex, parameters)::Dict
         checkIntStr = get(parameters, "amr.check_int", "")
         if checkIntStr != ""
             checkpointFrequency = parse(Int64, checkIntStr)
-            independentVariables["amr.ncheck_files"] = floor(Int32, maxStep / checkpointFrequency)
+            independentVariables["amr.check_int"] = checkpointFrequency
+            independentVariables["amr.ncheck_files"] =
+                floor(Int32, maxStep / checkpointFrequency)
         else
             throw(NoSuchFieldException("jexio Amrex: Can't find amr.check_int"))
         end
@@ -109,10 +106,15 @@ function get_independent_variables(extractor::AbstractAmrex, parameters)::Dict
 
     maxLevel = get(parameters, "amr.max_level", "")
     independentVariables["amr.max_level"] = maxLevel == "" ? 1 : parse(Int32, maxLevel)
-      
+
     regrid = get(parameters, "amr.regrid_int", "")
     independentVariables["amr.regrid_int"] = regrid == "" ? 1 : parse(Int32, regrid)
-    
+
     return independentVariables
 
 end
+
+
+function _get_inputs_X(extractor::AbstractAmrex)
+
+end

src/extractor/AmrexCastro.jl

 
 include("../helper/helperSystem.jl")
-  
 
-mutable struct AmrexCastro <: AbstractAmrex
+@time import DataFrames
+
+struct AmrexCastro <: AbstractAmrex
     app::String
-    degreesOfFreedom::Any
     outputPrefix::String
+    inputs::Array{String}
+    outputs::Dict{String,Array{String}}
 
-    AmrexCastro() = new()
+    AmrexCastro(outputPrefix::String) = new(
+        "AmrexCastro",
+        outputPrefix,
+        [
+            "max_step",
+            "amr.check_int",
+            "amr.plot_int",
+            "amr.n_cell",
+            "amr.max_level",
+            "castro.cfl",
+            "castro.max_grid_size",
+        ],
+        Dict{String,Array{String}}("plots_size" => ["max_step", "amr.plot_int", "amr.n_cell", "amr.max_level"])
+    )
 end
 
-"""
-   init!
-   initialize members of the extractor::AmrexCastro type
-   Using bang convention as init modifies the extractor::Amrex
-   https://docs.julialang.org/en/v1/manual/style-guide/index.html#bang-convention-1
-"""
-function init!(extractor::AmrexCastro, outputPrefix::String)
-    extractor.app = "AmrexCastro"
-    extractor.degreesOfFreedom = [
-        "max_step",
-        "amr.check_int",
-        "amr.plot_int",
-        "amr.n_cell",
-        "amr.max_level",
-        "castro.cfl",
-        "castro.max_grid_size",
-    ]
-    extractor.outputPrefix = outputPrefix
-end
 
+function _get_linear_model(extractor::AmrexCastro)
+    # get directories runs output location
+    outputDirs = helper_get_prefix_directories(extractor.outputPrefix)
+    # println(outputDirs)
+
+    X = DataFrames.DataFrame()
+    isXInit::Bool = false
+
+    for outputDir in outputDirs
+        inputFiles::Array{String} = helper_get_prefix_files("input", outputDir)
+
+        if isempty(inputFiles)
+            throw(NoSuchFieldException(
+                "jexio AmrexCastro: Can't find Castro input files in directories:",
+                outputDirs,
+            ))
+            # use length, instead of size as it returns tuples
+        elseif length(inputFiles) != 1
+            print("Number of files: ", size(inputFiles))
+            throw(DimensionMismatch("jexio AmrexCastro: found more than one input file"))
+        else
+        end
 
+        inputFile = inputFiles[1]
+
+        # get relevant input parameters
+        parameters = _input_parser(extractor, inputFile)
+        independent_variables = _get_independent_variables(extractor, parameters)
+
+        if !isXInit
+            X = DataFrames.DataFrame(independent_variables)
+            isXInit = true
+        else
+            DataFrames.push!(X, independent_variables)
+        end
+    end
+
+    println(X)
 
-function get_dependent_variable_datasize(extractor::AmrexCastro)
-   
-   outputDirs = helper_get_prefix_directories(extractor.outputPrefix)
-   println(outputDirs)
-  
 end
 
 
+function _get_outputs_Y(extractor::AmrexCastro)
 
+    function _get_outputs_Y_plots_size(extractor::AmrexCastro)
+        maxStepStr = get(extractor.parameters, "", "")
+    end
 
+    for output in extractor.outputs
+        if (output == "plots_size")
+            _get_outputs_Y_plots_size(extractor)
+        end
+    end
 
+end

src/helper/helperSystem.jl

@@ -3,11 +3,30 @@ import Glob
 
 
 function helper_get_prefix_directories(prefix::String)::Array{String}
-  findDelimiter = findlast("/",prefix)
-  searchDirectory = prefix[ 1: findDelimiter[1]-1 ]
-    print(searchDirectory)
-    
-  pattern =  string(prefix[ findDelimiter[1]+1:end],"*")
-  outputDirs = Glob.glob( pattern, searchDirectory )
-  return outputDirs
-end
\ No newline at end of file
+    findDelimiter = findlast("/", prefix)
+    searchDirectory::String = prefix[1:findDelimiter[1]-1]
+    #print(searchDirectory)
+
+    pattern::String = string(prefix[findDelimiter[1]+1:end], "*")
+    outputDirs = Glob.glob(pattern, searchDirectory)
+    return outputDirs
+end
+
+
+"""
+  Gets the list of files inside a path that meet the prefix criteria.
+  Example: 
+      path/prefix.*
+
+      path/
+           prefix.1
+           prefix.2
+           prefix.3
+
+  output: [ prefix.1, prefix.2, prefix.3 ] 
+"""
+function helper_get_prefix_files(prefix::String, path::String)::Array{String}
+
+    pattern::String = string(prefix, "*")
+    files = Glob.glob(pattern, path)
+end

test/runtests.jl

@@ -8,5 +8,5 @@ import Exio
 end;
 
 @testset "test_Exio.input_parser_docstring" begin
-    @test println(@doc Exio.input_parser) === nothing
+    @test println(@doc Exio._input_parser) === nothing
 end;

test/test_AmrexCastro.jl

@@ -16,18 +16,22 @@ function test_value(
 end
 
 function test_AmrexCastro()
-    exio = Exio.exio_init("AmrexCastro", "./test/data/AmrexCastro/hydro_tests/Sedov/2d.cyl_in_cartcoords/case")
+    exio = Exio.exio_init(
+        "AmrexCastro",
+        "./test/data/AmrexCastro/hydro_tests/Sedov/2d.cyl_in_cartcoords/case",
+    )
 
     @test typeof(exio) == Exio.ExioH
 
     println("Current directory: ", Filesystem.pwd())
-    inputFile = string(
-        Filesystem.pwd(), "/",
+    inputFile::String = string(
+        Filesystem.pwd(),
+        "/",
         "test/data/AmrexCastro/test_data_AmrexCastro_inputs.2d.cyl_in_cartcoords",
     )
     println("Input file: ", inputFile)
 
-    parameters = Exio.input_parser(exio.extractor, inputFile)
+    parameters = Exio._input_parser(exio.extractor, inputFile)
 
     # keys counter
     counter = 0
@@ -62,30 +66,23 @@ function test_AmrexCastro()
 
     @test counter == 27
 
-    independent_variables = Exio.get_independent_variables(exio.extractor, parameters)
+    independent_variables = Exio._get_independent_variables(exio.extractor, parameters)
     println(independent_variables)
-    
-    Exio.get_dependent_variable_datasize(exio.extractor)
+
+    Exio._get_linear_model(exio.extractor)
 
 end
 
 
 test_AmrexCastro()
 
-@time using DataFrames
-
-X=[1,2,3,4,5,6]
-Y=[2,4,7,9,11,13]
-data = DataFrames.DataFrame(X=X,Y=Y)
-println(data)
-
-@time using GLM
-
-ols = GLM.lm(@formula(Y ~ X), data)
-println(ols)
-
-@time using Plots
-display(plot(X,Y))
+X = [1, 2, 3, 4, 5, 6]
+Y = [2, 4, 7, 9, 11, 13]
 
+#@time import GLM
 
+#ols = GLM.lm(GLM.@formula(Y ~ X), data)
+# println(ols)
 
+@time import Plots
+display(Plots.plot(X, Y))