The Anomaly Detection component is used to detect data with continuous or enumeration features. It helps you detect exceptions in the data.
Background information
Anomalous features in data can be detected by using the Box Plot or Attribute Value Frequency (AVF) method.
- Box Plot is used to detect data with continuous features. The detection is performed based on the maximum and minimum values of the box plot chart.
- AVF is used to detect data with enumeration features. The detection is performed based on the frequency and threshold of enumeration features.
Configure the component
You can use one of the following methods to configure the Anomaly Detection component.
Method 1: Configure the component on the pipeline page
You can configure the parameters of the Anomaly Detection component on the pipeline page of Machine Learning Designer of Machine Learning Platform for AI (PAI). Machine Learning Designer is formerly known as Machine Learning Studio. The following table describes the parameters.
Tab | Parameter | Description |
---|---|---|
Fields Setting | Feature Columns | The fields to be analyzed. |
Anomaly Detection Method | The method used to detect anomalous data. Box Plot is used to detect data with continuous features. AVF is used to detect data with enumeration features. |
Method 2: Use PAI commands
Configure the component parameters by using PAI commands. You can use the SQL Script component to call PAI commands. For more information, see SQL Script.
PAI -name fe_detect_runner -project algo_public
-DselectedCols="emp_var_rate,cons_price_rate,cons_conf_idx,euribor3m,nr_employed" \
-Dlifecycle="28"
-DdetectStrategy="boxPlot"
-DmodelTable="pai_temp_2458_23565_2"
-DinputTable="pai_bank_data"
-DoutputTable="pai_temp_2458_23565_1";
Parameter | Description | Required |
---|---|---|
inputTable | The name of the input table. | Yes |
inputTablePartitions | The partitions in the input table. By default, all partitions are selected.
| No |
selectedCols | The input features. The data types of the features are not limited. | Yes |
detectStrategy | The detection method. Box Plot and AVF are supported. Box Plot is used to detect data with continuous features. AVF is used to detect data with enumeration features. | Yes |
outputTable | The output table that contains data with anomalous features. | Yes |
modelTable | The anomaly detection model. | Yes |
lifecycle | The lifecycle of the output table. Default value: 7. | No |
coreNum | The number of cores. This parameter must be used with the memSizePerCore parameter. Note The value of this parameter must be a positive integer. Valid values: 1 to 9999. | No |
memSizePerCore | The memory size of each core. Unit: MB. Valid values: [2048,64 × 1024]. | No |