diff --git a/buildsample/Build.DataMiningSample.cls b/buildsample/Build.DataMiningSample.cls index d45d27b..b67629d 100644 --- a/buildsample/Build.DataMiningSample.cls +++ b/buildsample/Build.DataMiningSample.cls @@ -38,12 +38,12 @@ ClassMethod Build() } /// This method enables you to build the sample with a minimum of output to the Terminal. -/// stagingroot is the name of the directory that contains the downloaded files. -/// interactive controls whether this method displays output. -/// Build calls this method with interactive=1. +/// stagingroot is the name of the directory that contains the downloaded files. +/// interactive controls whether this method displays output. +/// Build calls this method with interactive=1. ClassMethod run(stagingroot As %String = "", interactive As %Boolean = 0) { - //repeat this test here in case we go directly to this method + // Repeat this test here in case we go directly to this method if $ZSTRIP(stagingroot,"<>W")="" { if interactive { write !, "Not a valid directory... Quitting." @@ -51,7 +51,7 @@ ClassMethod run(stagingroot As %String = "", interactive As %Boolean = 0) return 0 } - //load and compile classes *************************** + // Load and compile classes *************************** set dir=stagingroot_"/cls/" ;works on both Windows and Unix if '##class(%File).DirectoryExists(dir) { if interactive { @@ -65,7 +65,7 @@ ClassMethod run(stagingroot As %String = "", interactive As %Boolean = 0) } do $system.OBJ.LoadDir(dir,"ck",,1) - // execute the setup code in the loaded classes ********************* + // Execute the setup code in the loaded classes ********************* if '##class(%Dictionary.CompiledClass).%ExistsId("DataMining.IrisDataset") { if interactive { write !!, "Looking for the class DataMining.IrisDataset" diff --git a/cls/DataMining/ClusterAnalysis/Iris.cls b/cls/DataMining/ClusterAnalysis/Iris.cls index fac0899..ab576d3 100644 --- a/cls/DataMining/ClusterAnalysis/Iris.cls +++ b/cls/DataMining/ClusterAnalysis/Iris.cls @@ -2,47 +2,43 @@ Include %occInclude /// Use or operation of this code is subject to acceptance of the license available in the code repository for this code. /// The test and demo for application of Cluster Analysis to Iris dataset contained in class -/// DataMining.IrisDataset -///

+/// DataMining.IrisDataset /// Run method ##class(DataMining.ClusterAnalysis.Iris).test() -///

/// See also Graphical Zen Demo implemented in class -/// DataMining.ClusterAnalysis.ZenDemoPage +/// DataMining.ClusterAnalysis.ZenDemoPage Class DataMining.ClusterAnalysis.Iris [ Abstract ] { /// This method runs the full Cluster Analysis procedure for Iris dataset. /// Parameters: -///

ClassMethod test(verbose As %Integer = 2, NClusters As %Integer = 3, method As %Integer = 1) { #dim sc As %Status = $$$OK #dim ex As %Exception.AbstractException #dim pam As %DeepSee.extensions.clusters.PAM - Try{ + try{ set pam = ..GetModel($s(method=2:"clara",method:"sa",1:"pam"), .sc) - Quit:$$$ISERR(sc) + quit:$$$ISERR(sc) set pam.K = NClusters set pam.Verbose = verbose - 1 set sc = pam.Execute() - Quit:$$$ISERR(sc) + quit:$$$ISERR(sc) - Quit:'verbose + quit:'verbose if (verbose > 2) { do pam.printAll() @@ -63,7 +59,7 @@ ClassMethod test(verbose As %Integer = 2, NClusters As %Integer = 3, method As % set sc = tPMML.%SaveToClass(tModelClassName, 1, verbose) quit:$$$ISERR(sc) - // run PMML model against the whole dataset + // Run PMML model against the whole dataset if (verbose) { set tClusterID = "" for i = 1:1:3 { @@ -81,7 +77,7 @@ ClassMethod test(verbose As %Integer = 2, NClusters As %Integer = 3, method As % quit:$$$ISERR(sc) } - }Catch(ex){ + }catch(ex){ set sc = ex.AsStatus() } do:$$$ISERR(sc) $system.OBJ.DisplayError(sc) @@ -97,9 +93,9 @@ ClassMethod speciesMatrix(model As %DeepSee.extensions.clusters.AbstractModel, k set i = "" set mostFrequent = "" - Do { + do { do model.iterateCluster(k, .i, .id, .x) - Quit:i="" + quit:i="" set flower = ##class(DataMining.IrisDataset).%OpenId(id) set species = flower.Species @@ -112,7 +108,7 @@ ClassMethod speciesMatrix(model As %DeepSee.extensions.clusters.AbstractModel, k } While(i'="") - Quit sc + quit sc } ClassMethod Summary(model As %DeepSee.extensions.clusters.AbstractModel, Output pClusterNames) As %Status @@ -127,7 +123,7 @@ ClassMethod Summary(model As %DeepSee.extensions.clusters.AbstractModel, Output k pClusterNames set sc = ..GetResults(model, .correct, .error, .s1, .s2, .s3) - Quit:$$$ISERR(sc) sc + quit:$$$ISERR(sc) sc set pClusterNames(1) = s1 set pClusterNames(2) = s2 @@ -141,11 +137,11 @@ ClassMethod Summary(model As %DeepSee.extensions.clusters.AbstractModel, Output w "Indices:", ! set asw = model.GetASWIndex().calculate(.sc) - Quit:$$$ISERR(sc) sc + quit:$$$ISERR(sc) sc set r = model.GetPearsonGammaIndex().calculate(.sc) - Quit:$$$ISERR(sc) sc + quit:$$$ISERR(sc) sc set ch = model.GetCalinskiHarabaszIndex().calculate(.sc) - Quit:$$$ISERR(sc) sc + quit:$$$ISERR(sc) sc set cost = model.TotalCost() w "Total Final Cost = "_cost, ! @@ -153,7 +149,7 @@ ClassMethod Summary(model As %DeepSee.extensions.clusters.AbstractModel, Output w "Pearson-Gamma (PG) = "_r, ! w "Calinski-Harabasz (CH) = "_ch, ! - Quit sc + quit sc } ClassMethod GetResults(model As %DeepSee.extensions.clusters.AbstractModel, Output correct As %Integer, Output error As %Integer, Output s1 As %String, Output s2 As %String, Output s3 As %String) @@ -162,20 +158,20 @@ ClassMethod GetResults(model As %DeepSee.extensions.clusters.AbstractModel, Outp #dim matrix set sc = ..speciesMatrix(model, 1, .matrix, .s1) - Quit:$$$ISERR(sc) sc + quit:$$$ISERR(sc) sc set sc = ..speciesMatrix(model, 2, .matrix, .s2) - Quit:$$$ISERR(sc) sc + quit:$$$ISERR(sc) sc set sc = ..speciesMatrix(model, 3, .matrix, .s3) - Quit:$$$ISERR(sc) sc + quit:$$$ISERR(sc) sc set correct = $g(matrix(1,s1)) + $g(matrix(2,s2)) + $g(matrix(3,s3)) set error = $g(matrix(1,s2),0) + $g(matrix(1,s3),0) set error = error + $g(matrix(2,s1),0) + $g(matrix(2,s3),0) set error = error + $g(matrix(3,s1),0) + $g(matrix(3,s2),0) - Quit sc + quit sc } ClassMethod GetModel(method As %String, Output sc As %Status) As %DeepSee.extensions.clusters.AbstractModel @@ -187,7 +183,7 @@ ClassMethod GetModel(method As %String, Output sc As %Status) As %DeepSee.extens if ('##class(%DeepSee.extensions.clusters.AbstractModel).Exists(dataset)){ set sc = ..PrepareModel(dataset) - Quit:$$$ISERR(sc) "" + quit:$$$ISERR(sc) "" } if (method="sa") { @@ -203,7 +199,7 @@ ClassMethod GetModel(method As %String, Output sc As %Status) As %DeepSee.extens set pam.Normalize = 0 - Quit pam + quit pam } ClassMethod checkData() As %Status @@ -211,19 +207,19 @@ ClassMethod checkData() As %Status #dim sc As %Status = $$$OK #dim ex As %Exception.AbstractException - Try{ + try{ &sql(select count(%ID) into :count from DataMining.IrisDataset) if (count'=150) { set sc = ##class(DataMining.IrisDataset).%KillExtent() - Quit:$$$ISERR(sc) + quit:$$$ISERR(sc) set sc = ##class(DataMining.IrisDataset).load() - Quit:$$$ISERR(sc) + quit:$$$ISERR(sc) } - }Catch(ex){ + }catch(ex){ set sc = ex.AsStatus() } - Quit sc + quit sc } ClassMethod PrepareModel(dataset As %String) As %Status @@ -232,35 +228,35 @@ ClassMethod PrepareModel(dataset As %String) As %Status #dim ex As %Exception.AbstractException #dim pam As %DeepSee.extensions.clusters.PAM - Try{ + try{ set sc = ..checkData() set pam = ##class(%DeepSee.extensions.clusters.PAM).New(dataset,.sc) - Quit:$$$ISERR(sc) + quit:$$$ISERR(sc) set pam.Normalize = 0 set rs = ##class(%ResultSet).%New("%DynamicQuery:SQL") set sc = rs.Prepare("select %ID, SepalLength, SepalWidth, PetalLength, PetalWidth from DataMining.IrisDataset") - Quit:$$$ISERR(sc) + quit:$$$ISERR(sc) set sc = rs.Execute() - Quit:$$$ISERR(sc) + quit:$$$ISERR(sc) set sc = pam.SetData(rs, 4) - Quit:$$$ISERR(sc) + quit:$$$ISERR(sc) set sc = pam.Prepare() - Quit:$$$ISERR(sc) - }Catch(ex){ + quit:$$$ISERR(sc) + }catch(ex){ set sc = ex.AsStatus() } do:$$$ISERR(sc) $system.OBJ.DisplayError(sc) - Quit sc + quit sc } ClassMethod DeleteModel() As %Status { - Quit ##class(%DeepSee.extensions.clusters.AbstractModel).Delete("iris") + quit ##class(%DeepSee.extensions.clusters.AbstractModel).Delete("iris") } } diff --git a/cls/DataMining/ClusterAnalysis/SimpleTest.cls b/cls/DataMining/ClusterAnalysis/SimpleTest.cls index 9e18535..d5d3aa9 100644 --- a/cls/DataMining/ClusterAnalysis/SimpleTest.cls +++ b/cls/DataMining/ClusterAnalysis/SimpleTest.cls @@ -1,11 +1,9 @@ /// Use or operation of this code is subject to acceptance of the license available in the code repository for this code. /// This is a simple demo and test for Cluster Analysis algorithms included with DeepSee. -///

/// The 2-dimensional dataset consists of 9 points clearly distributed between two clusters. -///

-/// Run ##class(DataMining.ClusterAnalysis.SimpleTest).test() -/// or see also Graphical Zen Demo implemented in class -/// DataMining.ClusterAnalysis.ZenDemoPage +/// Run ##class(DataMining.ClusterAnalysis.SimpleTest).test() +/// or see also Graphical Zen Demo implemented in class +/// DataMining.ClusterAnalysis.ZenDemoPage Class DataMining.ClusterAnalysis.SimpleTest Extends %Persistent { @@ -126,26 +124,5 @@ ClassMethod test(verbose As %Integer = 1, NClusters As %Integer = 2) do:$$$ISERR(sc) $system.OBJ.DisplayError(sc) } -Storage Default -{ - - -%%CLASSNAME - - -X - - -Y - - -^DataMining.Clus7500.SimpleTestD -SimpleTestDefaultData -^DataMining.Clus7500.SimpleTestD -^DataMining.Clus7500.SimpleTestI -^DataMining.Clus7500.SimpleTestS -%Library.CacheStorage -} - } diff --git a/cls/DataMining/IrisCube.cls b/cls/DataMining/IrisCube.cls index 9ed4230..9de3361 100644 --- a/cls/DataMining/IrisCube.cls +++ b/cls/DataMining/IrisCube.cls @@ -1,5 +1,4 @@ /// Use or operation of this code is subject to acceptance of the license available in the code repository for this code. -/// Class DataMining.IrisCube Extends %DeepSee.CubeDefinition [ DependsOn = DataMining.IrisDataset, ProcedureBlock ] { @@ -34,28 +33,28 @@ XData Cube [ XMLNamespace = "http://www.intersystems.com/deepsee" ] ClassMethod GetPredictedSpecies(pID As %String) As %String { - // execute the Decision Tree model to predict the species of the IrisDataset instance for pID + // Execute the Decision Tree model to predict the species of the IrisDataset instance for pID set tSC = $$$OK, tOutcome = "" try { - // 1: fetch the IrisDataset instance + // 1: Fetch the IrisDataset instance set tIris = ##class(DataMining.IrisDataset).%OpenId(pID) if (tIris="") { set tSC = $$$ERROR($$$GeneralError, "Could not open IrisDataset object with ID '"_pID_"'") quit } - // 2: load its properties into an array + // 2: Load its properties into an array kill tData set tData("PetalLength") = tIris.PetalLength set tData("PetalWidth") = tIris.PetalWidth set tData("SepalLength") = tIris.SepalLength set tData("SepalWidth") = tIris.SepalWidth - // 3: instantiate the Decision Tree model + // 3: Instantiate the Decision Tree model set tSC = ##class(DataMining.PMML.Iris).%GetModelInstance("DecisionTree", .tModel) quit:$$$ISERR(tSC) - // 4: execute the model and retrieve the predicted species + // 4: Execute the model and retrieve the predicted species set tSC = tModel.%ExecuteModel(.tData, .tOutput) quit:$$$ISERR(tSC) set tOutcome = tOutput.%GetFeatureValue("predictedValue") @@ -67,7 +66,7 @@ ClassMethod GetPredictedSpecies(pID As %String) As %String quit tOutcome } -/// Ensure DataMiningIrisDataset contains data +/// Ensure DataMiningIrisDataset contains data ClassMethod %OnBuildCube() As %Status { set tSC = ##class(DataMining.IrisDataset).%DeleteExtent() diff --git a/cls/DataMining/IrisDataset.cls b/cls/DataMining/IrisDataset.cls index a8a0735..9dd52df 100644 --- a/cls/DataMining/IrisDataset.cls +++ b/cls/DataMining/IrisDataset.cls @@ -1,20 +1,14 @@ /// Use or operation of this code is subject to acceptance of the license available in the code repository for this code. /// This class represents the Iris dataset, perhaps the best known database to be found in the pattern recognition literature. -///

/// The data set contains 3 classes of 50 instances each, where each class refers to a type of iris plant. One class is linearly separable from the other 2; the latter are NOT linearly separable from each other. Each record has 5 attributes: -///

    -///
  1. sepal length in cm -///
  2. sepal width in cm -///
  3. petal length in cm -///
  4. petal width in cm -///
  5. class (species): -/// -///
-///

+/// sepal length in cm +/// sepal width in cm +/// petal length in cm +/// petal width in cm +/// class (species): +/// Iris Setosa +/// Iris Versicolour +/// Iris Virginica /// The dataset is taken from UCI Machine Learning Repository [http://archive.ics.uci.edu/ml]. Irvine, CA: University of California, School of Information and Computer Science by Frank, A. & Asuncion, A. (2010). Class DataMining.IrisDataset Extends %Persistent { @@ -37,21 +31,21 @@ ClassMethod load() As %Status #dim line As %String #dim len, count As %Integer - Try { - Set input=##class(%Dictionary.CompiledXData).%OpenId("DataMining.IrisDataset||Iris").Data - If '$IsObject(input) Set sc=%objlasterror Quit + try { + set input=##class(%Dictionary.CompiledXData).%OpenId("DataMining.IrisDataset||Iris").Data + if '$IsObject(input) set sc=%objlasterror quit - Quit:$$$ISERR(sc) + quit:$$$ISERR(sc) - Set count = 0 - While ($$$ISOK(sc)) { + set count = 0 + while ($$$ISOK(sc)) { set len = 32000 set line = input.ReadLine(.len,.sc) - Quit:len<1 - Quit:$$$ISERR(sc) + quit:len<1 + quit:$$$ISERR(sc) - Continue:line["<" - Continue:line[">" + continue:line["<" + continue:line[">" set count = $i(count) @@ -63,10 +57,10 @@ ClassMethod load() As %Status set flower.Species = $p(line,",",5) set sc = flower.%Save() - Quit:$$$ISERR(sc) + quit:$$$ISERR(sc) } - } Catch (ex) { + } catch (ex) { set sc = ex.AsStatus() } @@ -74,7 +68,7 @@ ClassMethod load() As %Status do $system.OBJ.DisplayError(sc) } - Quit sc + quit sc } XData Iris @@ -234,36 +228,5 @@ XData Iris ]]> } - -Storage Default -{ - - -%%CLASSNAME - - -SepalLength - - -SepalWidth - - -PetalLength - - -PetalWidth - - -Species - - -^DataMining.IrisDatasetD -IrisDatasetDefaultData -^DataMining.IrisDatasetD -^DataMining.IrisDatasetI -^DataMining.IrisDatasetS -%Storage.Persistent -} - }