@frontlinesystems/rason-mcp-server 2026.5.0

package/examples/datamining/Transformation/CategoryReduction.json

@@ -0,0 +1,52 @@
+{
+  "modelName": "CategoryReduction",
+  "modelDescription": "Transformation RASON model example that performs category reduction on an imported data source. This datasource can be downloaded by clicking the Download RASON Example Files icon (download icon) on the ribbon. To attach the source file to the example, click Choose Files (on the Properties tab on the right), navigate to the location of the downloaded files, select hald-small.txt and click Open. Then click the POST icon (upload icon) to POST the model to the RASON Server, then click the down arrow next to the Play icon and select Solve to solve the model.",
+  "modelType": "datamining",
+  "datasources": {
+    "comment": "Imports data source, hald-small.txt, downloadable by clicking Download RASON example data icon, on the ribbon",
+    "mySrc": {
+      "comment": "Binds the hald-small.txt source file to the mySrc datasource.",
+      "type": "csv",
+      "connection": "hald-small.txt",
+      "direction": "import"
+    }
+  },
+  "datasets": {
+    "comment": "Datasets are bound to an entry in the datasources section",
+    "myData": {
+      "comment": "Binds mySrc datasource (hald-small.txt) to myData data set.",
+      "binding": "mySrc"
+    }
+  },
+  "transformer": {
+    "comment": "Defines a transformer, including the type, algorithm and parameters. Only 1 estimator/transformer may appear in a RASON model.",
+    "myReducer": {
+      "comment": "Defines the myReducer as a category reduction transformer. The number of categories are set for the X1 and X3 variables within parameters. For a complete list of all algorithm options, see the RASON Data Mining Components chapter in the RASON Reference Guide. (Click the Help tab to download.)",
+      "type": "transformation",
+      "algorithm": "categoryReduction",
+      "parameters": {
+        "numCategories": [
+          [ "X1", 3 ],
+          [ "X3", 2 ]
+        ]
+      }
+    }
+  },
+  "actions": {
+    "comment": "Define actions, evaluations, and data to be used by the transformer created above.",
+    "redModel": {
+      "comment": "Fits a category reduction model (using 'action':'fit') to the myData dataset ('data':'myData') using myReducer ('estimator':'myReducer'). See the RASON Reference Guide (click the Help tab to download) for a list of all supported category reduction parameters.",
+      "data": "myData",
+      "estimator": "myReducer",
+      "export": "json",
+      "action": "fit"
+    },
+    "reducedData": {
+      "comment": "Applies the fitted model (using 'action':'transform') to the myData dataset ('data':'myData') using the fitted category reduction model, redModel ('fittedModel':'redModel'). The only evaluation requested is the transformed data.",
+      "data": "myData",
+      "fittedModel": "redModel",
+      "action": "transform",
+      "evaluations": [ "transformation" ]
+    }
+  }
+}

package/examples/datamining/Transformation/Factorization.json

@@ -0,0 +1,51 @@
+{
+  "modelName": "Factorization",
+  "modelDescription": "Transformation RASON model example that performs factorization (category scores in Analytic Solver Data Mining) on an imported data source. This datasource can be downloaded by clicking the Download RASON Example Files icon (download icon) on the ribbon. To attach the source file to the example, click Choose Files (on the Properties tab on the right), navigate to the location of the downloaded files, select hald-small.txt and click Open. Then click the POST icon (upload icon) to POST the model to the RASON Server, then click the down arrow next to the Play icon and select Solve to solve the model.",
+  "modelType": "datamining",
+  "datasources": {
+    "comment": "Define datasources to use with this model. Typically these bind to external data sources due to data volume.",
+    "mySrc": {
+      "type": "csv",
+      "connection": "hald-small.txt",
+      "direction": "import"
+    }
+  },
+  "datasets": {
+    "comment": "Binds mySrc (hald-small.txt) datasource to myData dataset",
+    "myData": {
+      "comment": "Imports hald-small.txt datasource, downloadable by clicking Download RASON example data, on the ribbon",
+      "binding": "mySrc"
+    }
+  },
+  "estimator": {
+    "comment": "Defines an estimator, including the type, algorithm and parameters. Only 1 estimator/transformer may appear in a RASON model.",
+    "factorEstimator": {
+      "comment": "Defines factorEstimator as a factorization estimator. The base Index parameter is used to set the base index for the x1 and x3 variables.  For a complete list of all algorithm options, see the RASON Data Mining Components chapter in the RASON Reference Guide. (Click the Help tab to download.)",
+      "type": "transformation",
+      "algorithm": "factorization",
+      "parameters": {
+        "baseIndex": [
+          [ "X1", 1 ],
+          [ "X3", 5 ]
+        ]
+      }
+    }
+  },
+  "actions": {
+    "comment": "Define actions, evaluations, and data to be used by the estimator created above.",
+    "factorModel": {
+      "comment": "Fits a factorization model (using 'action':'fit') to the myData dataset ('data':'myData') using factorEstimator ('estimator':'factorEstimator'). The property export:json posts the fitted model to the RASON server. See the RASON Reference Guide (click the Help tab to download) for a list of all supported factorization parameters.",
+      "data": "myData",
+      "estimator": "factorEstimator",
+      "export": "json",
+      "action": "fit"
+    },
+    "factorizedData": {
+      "comment": "Applies the fitted model (using 'action':'transform') to the myData dataset ('data':'myData') using the factorization model, factorModel ('fittedModel':'factorModel'). Two evaluations are requested:  transformation (which returns the transformed dataset) and the frequency table.",
+      "data": "myData",
+      "fittedModel": "factorModel",
+      "action": "transform",
+      "evaluations": [ "transformation", "frequencyTable" ]
+    }
+  }
+}

package/examples/datamining/Transformation/MissingDataHandling.json

@@ -0,0 +1,84 @@
+{
+  "modelName": "MissingDataHandling",
+  "modelDescription": "Transformation RASON model example that performs imputation (or missing data handling).   Click the POST icon (upload icon) to POST the model to the RASON Server, then click the down arrow next to the Play icon and select Solve to solve the model.",
+  "modelType": "datamining",
+  "datasets": {
+    "comment": "Data is provided inline so no need for the datasource section.",
+    "trainData": {
+      "comment": "Represents the training partition. The colNames property provides the column headings for the data passed for value.",
+      "value": [
+        [ "black", null, 6.0, 2.0, 1.0, "nan", 1 ],
+        [ "", 3.0, 9.0, 5.1, null, "", 2 ],
+        [ "red", 7.0, 8.0, null, 9.2, "small", 3 ],
+        [ "red", 10000.0, null, 4.4, 4.4, "large", -1 ],
+        [ "blue", 2, 3, 5.6, 3.4, "unknown", 5 ]
+      ],
+      "colNames": [ "A", "B", "C", "D", "E", "F", "G" ]
+    },
+    "newData": {
+      "comment": "Represents new data. The colNames property provides the column headings for the data passed for value.",
+      "value": [
+        [ "blue", null, null, 4.0, 1.0, "unknown", -1 ]
+      ],
+      "colNames": [ "A", "B", "C", "D", "E", "F", "G" ]
+    }
+  },
+  "estimator": {
+    "comment": "Defines an estimator, including the type, algorithm and parameters. Only 1 estimator/transformer may appear in a RASON model.",
+    "myImputer": {
+      "comment": "Defines myImputer as a imputation (missing data handling) estimator. The imputation strategies for each column (variable) are set using the imputationStrategy parameter. The placeholder parameter sets the custom placeholder for missing values in a column. For example, if a missing value is found in the A column, the mode of the column will be inserted for the missing value, if a value of 10000.0 is found in the B column, the mean of the column will replace the 10000.0 vaue, if a missing value is found in the C column, the column median will be inserted for the missing value, if a missing value is found in the D or E columns, the record will be deleted, if a missing value is found in the F column, a user defined value will be entered, if a missing value is found in the G column, the column mean will be inserted for the missing value. See the imputation model parameter for the user defined value for the F column (i.e. see the parameter for fixedTrainData within actions, below. For a complete list of all algorithm options, see the RASON Data Mining Components chapter in the RASON Reference Guide. (Click the Help tab to download.)",
+      "type": "transformation",
+      "algorithm": "imputation",
+      "parameters": {
+        "imputationStrategy": [
+          [ "A", "MODE" ],
+          [ "B", "MEAN" ],
+          [ "C", "MEDIAN" ],
+          [ "D", "DELETE_RECORD" ],
+          [ "E", "DELETE_RECORD" ],
+          [ "F", "VALUE" ],
+          [ "G", "MEAN" ]
+        ],
+        "placeholder": [
+          [ "B", 10000.0 ],
+          [ "F", "unknown" ],
+          [ "G", -1 ]
+        ]
+      }
+    }
+  },
+  "actions": {
+    "comment": "Define actions, evaluations, and data to be used by the estimator created above.",
+    "imputingModel": {
+      "comment": "Fits an imputation model (using 'action':'fit') to the myData dataset ('data':'myData') using myImputer ('estimator':'myImputer'). The property export:json posts the fitted model to the RASON server using the same name as this example. See the RASON Reference Guide (click the Help tab to download) for a list of all supported factorization parameters.",
+      "data": "trainData",
+      "estimator": "myImputer",
+      "action": "fit",
+      "export": "json"
+    },
+    "fixedTrainData": {
+      "comment": "Applies the fitted model (using 'action':'transform') to the myData dataset ('data':'myData') using the imputation model, imputingModel ('fittedModel':'imputingModel'). The imputation model parameter specifies the user defined value to be inserted for a missing value in the F column. Three evaluations are requested: transformation, the dataset with imputation applied, the records with missing values and the records to delete. See the JSON results for the results of recordswithMissingValues and recordsToDelete (i.e. click Show JSON results in the upper right hand coner of the ODATA results viewer.)",
+      "data": "trainData",
+      "fittedModel": "imputingModel",
+      "action": "transform",
+      "parameters": {
+        "imputation": [
+          [ "F", "medium" ]
+        ]
+      },
+      "evaluations": [ "transformation", "recordsWithMissingValues", "recordsToDelete" ]
+    },
+    "fixedNewData": {
+      "comment": "Applies the fitted model (using 'action':'transform') to the newData dataset ('data':'newData') using the imputation model, imputingModel ('fittedModel':'imputigModel'). Three evaluations are requested:  transformation (which returns the transformed data), records with Missing Values and the deleted records. See the JSON results for the results of recordswithMissingValues and recordsToDelete (i.e. click Show JSON results in the upper right hand coner of the ODATA results viewer.)",
+      "data": "newData",
+      "fittedModel": "imputingModel",
+      "action": "transform",
+      "parameters": {
+        "imputation": [
+          [ "C", 7.5 ]
+        ]
+      },
+      "evaluations": [ "transformation", "recordsWithMissingValues", "recordsToDelete" ]
+    }
+  }
+}

package/examples/datamining/Transformation/OneHotEncoding-DataInline.json

@@ -0,0 +1,55 @@
+{
+  "modelName": "OneHotEncodingInlineData",
+  "modelDescription": "Transformation RASON model example that performs one-hot encoding (Create Dummies in Analytic Solver). In this example, the data is supplied inline. Click the POST icon (upload icon) to POST the model to the RASON Server, then click the down arrow next to the Play icon and select Solve to solve the model.",
+  "modelType": "datamining",
+  "datasets": {
+    "comment": "Binds the data supplied in the example to the 2 different datasets:  trainData and newData.",
+    "trainData": {
+      "comment": "Binds the data to the trainData dataset.",
+      "value": [
+        [ "1", 2.0 ],
+        [ "0", 3.0 ],
+        [ "1", 6.0 ]
+      ]
+    },
+    "newData": {
+      "comment": "Binds the data to the newData dataset.",
+      "value": [
+        [ "0", 3.0 ],
+        [ "1", 6.0 ],
+        [ "1", 6.0 ]
+      ]
+    }
+  },
+  "estimator": {
+    "comment": "Defines an estimator, including the type, algorithm and parameters. Only 1 estimator may appear in a RASON model.",
+    "myEncoder": {
+      "comment": "Defines myEncoder as a one hot encoder (Create Dummies) estimator. For a complete list of all algorithm options, see the RASON Data Mining Components chapter in the RASON Reference Guide. (Click the Help tab to download.)",
+      "type": "transformation",
+      "algorithm": "oneHotEncoding"
+    }
+  },
+  "actions": {
+    "comment": "Define actions, evaluations, and data to be used by the estimator created above.",
+    "encoderModel": {
+      "comment": "Fits (using 'action':'fit') a model to the trainData dataset (using 'data':'trainData') using one hot encoder estimator ('estimator':'myEncoder').",
+      "trainData": "trainData",
+      "estimator": "myEncoder",
+      "action": "fit"
+    },
+    "encodedTrainData": {
+      "comment": "Applies the fitted model (using 'action':'transform') to the trainData dataset ('data':'trainData') using the one hot encoding model, encoderModel ('fittedModel':'encoderModel'). The transformed trainData dataset will be returned in the JSON results.",
+      "data": "trainData",
+      "fittedModel": "encoderModel",
+      "action": "transform",
+      "evaluations": [ "transformation" ]
+    },
+    "encodedNewData": {
+      "comment": "Applies the fitted model (using 'action':'transform') to the trainData dataset ('data':'newData') using the one hot encoding model, encoderModel ('fittedModel':'encoderModel'). The transformed newData dataset will be returned in the JSON results.",
+      "data": "newData",
+      "fittedModel": "encoderModel",
+      "action": "transform",
+      "evaluations": [ "transformation" ]
+    }
+  }
+}

package/examples/datamining/Transformation/OneHotEncoding-DatainFile.json

@@ -0,0 +1,49 @@
+{
+  "modelName": "OneHotEncodingFileData",
+  "modelDescription": "Transformation RASON model example that performs one-hot encoding (Create Dummies in Analytic Solver). This example uses an outside data source, hald-small.txt, which can be downloaded by clicking the Download RASON Example Files icon (download icon) on the ribbon. To attach the source file to the example, click Choose Files (on the Properties tab on the right), navigate to the location of the downloaded files, select hald-small.txt and click Open. Click the POST icon (upload icon) to POST the model to the RASON Server, then click the down arrow next to the Play icon and select Solve to solve the model.",
+  "modelType": "datamining",
+  "datasources": {
+    "comment": "Define datasources to use with this model. Typically these bind to external data sources due to data volume.",
+    "mySource": {
+      "comment": "Imports hald-small.txt as datasource, downloadable by clicking Download RASON example data icon, on the ribbon",
+      "type": "csv",
+      "connection": "hald-small.txt",
+      "direction": "import"
+    }
+  },
+  "datasets": {
+    "comment": "Datasets are bound to an entry in the datasources section",
+    "myData": {
+      "comment": "Binds the mySource datasource (hald-small.txt) to the myData dataset.",
+      "binding": "mySource"
+    }
+  },
+  "estimator": {
+    "comment": "Defines an estimator, including the type, algorithm and parameters. Only 1 estimator/transformer may appear in a RASON model.",
+    "myEncoder": {
+      "comment": "Defines myEncoder as a one hot encoder (Create Dummies) estimator; the categorical features are selected using parameters. For a complete list of all algorithm options, see the RASON Data Mining Components chapter in the RASON Reference Guide. (Click the Help tab to download.)",
+      "type": "transformation",
+      "algorithm": "oneHotEncoding",
+      "parameters": {
+        "categoricalFeaturesNames": [ "X1", "X2", "X4" ]
+      }
+    }
+  },
+  "actions": {
+    "comment": "Define actions, evaluations, and data to be used by the estimator created above.",
+    "encoderModel": {
+      "comment": "Fits (using 'action':'fit') a model to the myData dataset (using 'data':'myData') using one hot encoder estimator ('estimator':'myEncoder'), POSTS the fitted model (encoderModel) to the RASON Server, in json format, using export:json.",
+      "trainData": "myData",
+      "estimator": "myEncoder",
+      "export": "json",
+      "action": "fit"
+    },
+    "encodedTrainData": {
+      "comment": "Applies the fitted model (using 'action':'transform') to the myData dataset ('data':'myData') using the one hot encoding model, encoderModel ('fittedModel':'encoderModel'). The transformed myData dataset will be returned in the results.",
+      "data": "myData",
+      "fittedModel": "encoderModel",
+      "action": "transform",
+      "evaluations": [ "transformation" ]
+    }
+  }
+}

package/examples/datamining/Transformation/PartitionTimeSeries.json

@@ -0,0 +1,49 @@
+{
+  "modelName": "TimeSeriesPartitioning",
+  "modelDescription": "Transformation RASON model example that performs partitioning on a time series dataset. This example uses an outside data source, Airpass.txt, which can be downloaded by clicking the Download RASON Example Files icon (download icon) on the ribbon. To attach the source file to the example, click Choose Files (on the Properties tab on the right), navigate to the location of the downloaded files, select airpass.txt and click Open. Click the POST icon (upload icon) to POST the model to the RASON Server, then click the down arrow next to the Play icon and select Solve to solve the model.",
+  "modelType": "datamining",
+  "datasources": {
+    "comment": "Define datasources to use with this model. Typically these bind to external data sources due to data volume.",
+    "airpass_src": {
+      "comment": "Imports Airpass.txt as datasource, downloadable by clicking Download RASON example data icon, on the ribbon. The property content:time-series considers the 1st column as the time variable and the 2nd column as the value column in the dataframe.",
+      "type": "csv",
+      "connection": "Airpass.txt",
+      "direction": "import",
+      "content": "time-series"
+    }
+  },
+  "datasets": {
+    "comment": "Datasets are bound to an entry in the datasources section",
+    "airpass_dataset": {
+      "comment": "SelectedCols optional; use if more than 2 columns in dataset, 1st selected col is the time variable and 2nd selected col is the value column",
+      "binding": "airpass_src",
+      "selectedCols": [ "Month", "Passengers" ]
+    }
+  },
+  "transformer": {
+    "comment": "Defines a transformer, including the type, algorithm and parameters. Only 1 estimator/transformer may appear in a RASON model.",
+    "myPartitioner": {
+      "comment": "Defines myPartitioner as a time series partitioning transformer. The partitioning ratios and random seed options are set in parameters. Note that in time series partitioning, the records are assigned to different partitions in sequential order. For a complete list of all algorithm options, see the RASON Data Mining Components chapter in the RASON Reference Guide. (Click the Help tab to download.)",
+      "type": "transformation",
+      "algorithm": "partitioning",
+      "parameters": {
+        "comment": "Use partitionMethod: Sequential for time series partition",
+        "partitionMethod": "SEQUENTIAL",
+        "ratios": [
+          [ "Training", 0.6 ],
+          [ "Validation", 0.4 ]
+        ],
+        "seed": 123
+      }
+    }
+  },
+  "actions": {
+    "comment": "Define actions, evaluations, and data to be used by the transformer created above.",
+    "partitions": {
+      "comment": "Applies the time series partitioning time series estimator, myPartitioner, to partition the time series dataset, airpass_dataset. Note that when performing a time series partition, the first 60% of records in the dataset will be sequentially assigned to the training partition and the remaining 40% of records will be assigned to the validation partition. For a complete list of all parameters and evaluations supported by time series partitioning, see the RASON Data Mining Components chapter in the RASON Reference Guide. ",
+      "data": "airpass_dataset",
+      "action": "transform",
+      "evaluations": [ "transformation" ]
+    }
+  }
+}

package/examples/datamining/Transformation/Partitioning-DF.json

@@ -0,0 +1,73 @@
+{
+  "modelName": "PartitioningDF",
+  "modelDescription": "transformation: partitioning",
+  "modelType": "datamining",
+  "datasources": {
+    "mySrc": {
+      "type": "csv",
+      "connection": "hald-small-binary.txt",
+      "direction": "import"
+    }
+  },
+  "datasets": {
+    "myData": {
+      "binding": "mySrc"
+    }
+  },
+  "transformer": {
+    "myPartitioner": {
+      "type": "transformation",
+      "algorithm": "partitioning",
+      "parameters": {
+        "partitionMethod": "RANDOM",
+        "ratios": [
+          [
+            "Training",
+            0.5
+          ],
+          [
+            "Validation",
+            0.3
+          ],
+          [
+            "Testing",
+            0.2
+          ]
+        ],
+        "seed": 123
+      }
+    }
+  },
+  "actions": {
+    "trainingPartition": {
+      "data": "myData",
+      "action": "transform",
+      "parameters": {
+        "partition": "Training"
+      },
+      "evaluations": [
+        "transformation"
+      ]
+    },
+    "validationPartition": {
+      "data": "myData",
+      "action": "transform",
+      "parameters": {
+        "partition": "Validation"
+      },
+      "evaluations": [
+        "transformation"
+      ]
+    },
+    "testPartition": {
+      "data": "myData",
+      "action": "transform",
+      "parameters": {
+        "partition": "Testing"
+      },
+      "evaluations": [
+        "transformation"
+      ]
+    }
+  }
+}

package/examples/datamining/Transformation/Partitioning.json

@@ -0,0 +1,47 @@
+{
+  "modelName": "Partitioning",
+  "modelDescription": "Transformation RASON model example that performs partitioning. This example uses an outside data source, hald-small-binary.txt, which can be downloaded by clicking the Download RASON Example Files icon (download icon) on the ribbon. To attach the source file to the example, click Choose Files (on the Properties tab on the right), navigate to the location of the downloaded files, select hald-small-binary.txt and click Open. Click the POST icon (upload icon) to POST the model to the RASON Server, then click the down arrow next to the Play icon and select Solve to solve the model.",
+  "modelType": "datamining",
+  "datasources": {
+    "comment": "Define datasources to use with this model. Typically these bind to external data sources due to data volume.",
+    "mySrc": {
+      "comment": "Imports hald-small-binary.txt as datasource, downloadable by clicking Download RASON example data icon, on the ribbon",
+      "type": "csv",
+      "connection": "hald-small-binary.txt",
+      "direction": "import"
+    }
+  },
+  "datasets": {
+    "comment": "Datasets are bound to an entry in the datasources section",
+    "myData": {
+      "comment": "Binds the mySrc datasource (hald-small-binary.txt) to the myData dataset.",
+      "binding": "mySrc"
+    }
+  },
+  "transformer": {
+    "comment": "Defines a transformer, including the type, algorithm and parameters. Only 1 estimator/transformer may appear in a RASON model.",
+    "myPartitioner": {
+      "comment": "Defines myPartitioner as a partitioning transformer. The type of partition (standard), the random seed setting, and the percentages of records to be assigned to the training, validation and testing partitions are set in parameters. For a complete list of all algorithm options, see the RASON Data Mining Components chapter in the RASON Reference Guide. (Click the Help tab to download.)",
+      "type": "transformation",
+      "algorithm": "partitioning",
+      "parameters": {
+        "partitionMethod": "RANDOM",
+        "ratios": [
+          [ "Training", 0.5 ],
+          [ "Validation", 0.3 ],
+          [ "Testing", 0.2 ]
+        ],
+        "seed": 123
+      }
+    }
+  },
+  "actions": {
+    "comment": "Define actions, evaluations, and data to be used by the transformer created above.",
+    "partitions": {
+      "comment": "Applies the transformer (using 'action':'transform') to the myData dataset ('data':'myData') using the mypartitioner transformer. The only evaluation requested is the transformed dataset, i.e. dataset divided into three partitions.",
+      "data": "myData",
+      "action": "transform",
+      "evaluations": [ "transformation" ]
+    }
+  }
+}

package/examples/datamining/Transformation/PartitioningWithOversampling-DF.json

@@ -0,0 +1,66 @@
+{
+  "modelName": "PartitioningWithOversamplingDF",
+  "modelDescription": "Transformation RASON model example that performs partitioning with oversampling. The difference between this oversampling example and the PartitionWithOversampling example is that this example returns the partitions as dataframes. This example uses an outside data source, hald-small-binary.txt, which can be downloaded by clicking the Download RASON Example Files icon (download icon) on the ribbon. To attach the source file to the example, click Choose Files (on the Properties tab on the right), navigate to the location of the downloaded files, select hald-small-binary.txt and click Open. Click the POST icon (upload icon) to POST the model to the RASON Server, then click the down arrow next to the Play icon and select Solve to solve the model.",
+  "modelType": "datamining",
+  "datasources": {
+    "comment": "Define datasources to use with this model. Typically these bind to external data sources due to data volume.",
+    "mySrc": {
+      "comment": "Imports hald-small-binary.txt as datasource, downloadable by clicking Download RASON example data icon, on the ribbon",
+      "type": "csv",
+      "connection": "hald-small-binary.txt",
+      "direction": "import"
+    }
+  },
+  "datasets": {
+    "comment": "Datasets are bound to an entry in the datasources section",
+    "myData": {
+      "comment": "Binds the mySrc datasource (hald-small-binary.txt) to the myData dataset.",
+      "binding": "mySrc",
+      "targetCol": "Y"
+    }
+  },
+  "transformer": {
+    "comment": "Defines a transformer, including the type, algorithm and parameters. Only 1 estimator/transformer may appear in a RASON model.",
+    "myPartitioner": {
+      "comment": "Defines myPartitioner as a partitioning using oversampling transformer. Four options are set in parameters. The parameter successRationinTraining sets the percentage of successes to be assigned to the training set. With this setting, the default, 50% of the successes will be assigned to the training partition and the remainder will be assigned to the validation partition. The last parameter, testRAtioFromValidation, specifies the percentage of the validation set that should be allocated to the test set. For a complete list of all algorithm options, see the RASON Data Mining Components chapter in the RASON Reference Guide. (Click the Help tab to download.)",
+      "type": "transformation",
+      "algorithm": "oversamplePartitioning",
+      "parameters": {
+        "seed": 12345,
+        "successClass": "1",
+        "successRatioInTraining": 0.5,
+        "testRatioFromValidation": 0.1
+      }
+    }
+  },
+  "actions": {
+    "comment": "Define actions, evaluations, and data to be used by the transformer created above.",
+    "trainingPartition": {
+      "comment": "Applies the transformer (using 'action':'transform') to the myData dataset ('data':'myData') using the mypartitioner transformer to create the Training partition. The only evaluation requested is the training partition which will be returned as a dataframe.",
+      "data": "myData",
+      "action": "transform",
+      "parameters": {
+        "partition": "Training"
+      },
+      "evaluations": [ "transformation" ]
+    },
+    "validationPartition": {
+      "comment": "Applies the transformer (using 'action':'transform') to the myData dataset ('data':'myData') using the mypartitioner transformer to create the Validation partition. The only evaluation requested is the validation partition which will be returned as a dataframe..",
+      "data": "myData",
+      "action": "transform",
+      "parameters": {
+        "partition": "Validation"
+      },
+      "evaluations": [ "transformation" ]
+    },
+    "testPartition": {
+      "comment": "Applies the transformer (using 'action':'transform') to the myData dataset ('data':'myData') using the mypartitioner transformer to create the Test partition. The only evaluation requested is the test partition which will be returned as a dataframe..",
+      "data": "myData",
+      "action": "transform",
+      "parameters": {
+        "partition": "Testing"
+      },
+      "evaluations": [ "transformation" ]
+    }
+  }
+}

package/examples/datamining/Transformation/PartitioningWithOversampling.json

@@ -0,0 +1,45 @@
+{
+  "modelName": "PartitioningWithOversampling",
+  "modelDescription": "Transformation RASON model example that performs partitioning with oversampling. The difference between this oversampling example and the PartitionWithOversamplingDF example is that this example does not return the partitions as dataframes. This example uses an outside data source, hald-small-binary.txt, which can be downloaded by clicking the Download RASON Example Files icon (download icon) on the ribbon. To attach the source file to the example, click Choose Files (on the Properties tab on the right), navigate to the location of the downloaded files, select hald-small-binary.txt and click Open. Click the POST icon (upload icon) to POST the model to the RASON Server, then click the down arrow next to the Play icon and select Solve to solve the model.",
+
+  "modelType": "datamining",
+  "datasources": {
+    "comment": "Define datasources to use with this model. Typically these bind to external data sources due to data volume.",
+    "mySrc": {
+      "comment": "Imports hald-small-binary.txt as datasource, downloadable by clicking Download RASON example data icon, on the ribbon",
+      "type": "csv",
+      "connection": "hald-small-binary.txt",
+      "direction": "import"
+    }
+  },
+  "datasets": {
+    "comment": "Datasets are bound to an entry in the datasources section",
+    "myData": {
+      "comment": "Binds the mySrc datasource (hald-small-binary.txt) to the myData dataset.",
+      "binding": "mySrc",
+      "targetCol": "Y"
+    }
+  },
+  "transformer": {
+    "myPartitioner": {
+      "comment": "Defines myPartitioner as a partitioning using oversampling transformer. Four options are set in parameters. The parameter successRationinTraining sets the percentage of successes to be assigned to the training set. With this setting, the default, 50% of the successes will be assigned to the training partition and the remainder will be assigned to the validation partition. The last parameter, testRAtioFromValidation, specifies the percentage of the validation set that should be allocated to the test set. For a complete list of all algorithm options, see the RASON Data Mining Components chapter in the RASON Reference Guide. (Click the Help tab to download.)",
+      "type": "transformation",
+      "algorithm": "oversamplePartitioning",
+      "parameters": {
+        "seed": 12345,
+        "successClass": "1",
+        "successRatioInTraining": 0.5,
+        "testRatioFromValidation": 0.1
+      }
+    }
+  },
+  "actions": {
+    "comment": "Define actions, evaluations, and data to be used by the transformer created above.",
+    "partitions": {
+      "comment": "Applies the transformer (using 'action':'transform') to the myData dataset ('data':'myData') using the mypartitioner transformer to create all three partitions using oversampling. The only evaluation requested is the are the three partitions.",
+      "data": "myData",
+      "action": "transform",
+      "evaluations": [ "transformation" ]
+    }
+  }
+}

package/examples/datamining/Transformation/PrincipalComponentsAnalysis.json

@@ -0,0 +1,53 @@
+{
+  "modelName": "PrincipalComponentsAnalysis",
+  "modelDescription": "Transformation RASON model example that performs principal components analysis. This example uses an outside data source, irisfacto.txt, which can be downloaded by clicking the Download RASON Example Files icon (download icon) on the ribbon. To attach the source file to the example, click Choose Files (on the Properties tab on the right), navigate to the location of the downloaded files, select irisfacto.txt and click Open. Click the POST icon (upload icon) to POST the model to the RASON Server, then click the down arrow next to the Play icon and select Solve to solve the model.",
+  "modelType": "datamining",
+  "comment": "Imports data source, downloadable by clicking Download RASON example model icon, on the ribbon",
+  "datasources": {
+    "comment": "Imports Irisfacto.txt as datasource, downloadable by clicking Download RASON example data icon, on the ribbon",
+    "myDataSrc": {
+      "type": "csv",
+      "connection": "Irisfacto.txt",
+      "direction": "import"
+    }
+  },
+  "datasets": {
+    "comment": "Datasets are bound to an entry in the datasources section",
+    "myData": {
+      "comment": "Binds the myDataSrc datasource (Irisfacto.txt) to the myData dataset.",
+      "binding": "myDataSrc"
+    }
+  },
+  "estimator": {
+    "comment": "Defines an estimator, including the type, algorithm and parameters. Only 1 estimator/transformer may appear in a RASON model.",
+    "pcaEstimator": {
+      "comment": "Defines pcaEstimator as a principal components analysis estimator. The correlation parameter is set in parameters. For a complete list of all algorithm options, see the RASON Data Mining Components chapter in the RASON Reference Guide. (Click the Help tab to download.)",
+      "type": "transformation",
+      "algorithm": "principalComponentAnalysis",
+      "parameters": {
+        "matrixMethod": "CORRELATION"
+      }
+    }
+  },
+  "actions": {
+    "comment": "Define actions, evaluations, and data to be used by the transformer created above.",
+    "pcaModel": {
+      "comment": "Fits (using 'action':'fit') a model to the myData dataset (using 'data':'myData') using the principal component analysis estimator ('estimator':'pcaEstimator'), POSTS the fitted model (pcaModel) to the RASON Server, in json format, using export:json. The fitted model will be returned in the JSON results. (Click Show JSON results, upper right hand corner of OData viewer, to view the fitted model.)",
+      "data": "myData",
+      "estimator": "pcaEstimator",
+      "action": "fit",
+      "export": "json",
+      "evaluations": [ "fittedModelJson" ]
+    },
+    "transformedData": {
+      "comment": "Applies the fitted model (using 'action':'transform') to the myData dataset ('data':'myData') using the PCA model, pcaModel ('fittedModel':'pcaModel'). Various evaluations, including the transformed data, will be returned in the results. For a complete list of all algorithm options, see the RASON Data Mining Components chapter in the RASON Reference Guide. (Click the Help tab to download.)",
+      "data": "myData",
+      "fittedModel": "pcaModel",
+      "parameters": {
+        "numPrincipalComponents": 3
+      },
+      "action": "transform",
+      "evaluations": [ "transformation", "principalEigenvalues", "principalComponents", "principalVariance", "qStatistic", "tSquaredStatistic" ]
+    }
+  }
+}