Scorecard Prediction - Platform For AI - Alibaba Cloud Documentation Center

Scorecard prediction is a machine learning technique that involves applying a scorecard model to new data to predict its future performance or risk. This model is typically generated by a scorecard training component, while the scorecard prediction component uses the model to evaluate and rate input data, thereby aiding in decision-making and risk management.

Configure the component

You can use one of the following methods to configure the Scorecard Prediction component.

Method 1: Configure the component on the pipeline page

You can configure the parameters of the Scorecard Prediction 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 feature columns that are used in prediction. By default, all feature columns are selected.
	Reserved Columns	The columns that are appended to the prediction result table without processing, such as the ID and objective columns.
	Output Variable Score	Specifies whether to generate a score for each feature variable. The total predicted score is the score of intercept options plus the score of each variable.
Tuning	Cores	The number of CPU cores that are required. By default, the system determines the value.
Tuning	Memory Size per Core	The memory size of each CPU core. By default, the system determines the value.

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=lm_predict
    -project=algo_public
    -DinputFeatureTableName=input_data_table
    -DinputModelTableName=input_model_table
    -DmetaColNames=sample_key,label
    -DfeatureColNames=fea1,fea2
    -DoutputTableName=output_score_table

Parameter	Description	Required	Default value
inputFeatureTableName	The name of the input feature table.	Yes	No default value
inputFeatureTablePartitions	The partitions that are selected from the input feature table.	No	Full table
inputModelTableName	The name of the input model table.	Yes	No default value
featureColNames	The feature columns that are selected from the input table.	No	All columns
metaColNames	The columns that do not need to be converted. These columns in the output are the same as those in the input. You can specify labels and sample IDs in the columns.	No	No default value
outputFeatureScore	Specifies whether to generate the scores of variables in the prediction results. Valid values: true: generates the scores of variables in the prediction results. false: does not generate the scores of variables in the prediction results.	No	false
outputTableName	The name of the output table.	Yes	No default value
lifecycle	The lifecycle of the output table.	No	No default value
coreNum	The number of cores.	No	Determined by the system
memSizePerCore	The memory size of each core. Unit: MB.	No	Determined by the system

Output

The following figure shows a score table generated by the Scorecard Prediction component. Sample score table The churn column is the column appended to the result table from the input table. The data in this column does not affect the prediction results. The other three columns display the prediction results. The following table describes the three columns.

Column	Data type	Description
prediction_score	DOUBLE	The column that contains predicted scores. In a linear model, the feature values and model weight values are multiplied and summed up to obtain the predicted scores. In a scorecard model, if score transformation is performed, the transformed scores are generated in this column.
prediction_prob	DOUBLE	The column that contains the probability values of positive samples in binary classification. The probability values are transformed from the original scores (without score transformation) by using the sigmoid function.
prediction_detail	STRING	The column that contains the probability values of positive and negative samples described in JSON strings. The value 0 represents negative, and the value 1 represents positive. Example: {"0":0.1813110520,"1":0.8186889480}.