Regression Model Evaluation

The Regression Model Evaluation component is used to evaluate the advantages and disadvantages of the different models of regression algorithms based on prediction results and original results. Then, evaluation metrics and histograms of residuals are generated.

You can use one of the following methods to configure the Regression Model Evaluation component.

Method 1: Configure the component on the pipeline page

You can configure the parameters of the Regression Model Evaluation 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	Original Regression Value	The columns of numeric data types are supported.
Fields Setting	Predicted Regression Value	The columns of numeric data types are supported.
Tuning	Worker number	The number of cores. Valid values: 1 to 9999. This parameter must be used with the Memory Size per Node parameter.
Tuning	Memory Size per Node	The memory size of each core. Valid values: 1024 to 64 × 1024. Unit: MB.

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 regression_evaluation -project algo_public
    -DinputTableName=input_table
    -DyColName=y_col
    -DpredictionColName=prediction_col
    -DindexOutputTableName=index_output_table
    -DresidualOutputTableName=residual_output_table;


Parameter	Required	Description	Default value
inputTableName	Yes	The name of the input table.	N/A
inputTablePartitions	No	The partitions that are selected from the input table for computing.	Full table
yColName	Yes	The name of the column that contains original dependent variables in the input table. The columns of numeric data types are supported.	N/A
predictionColName	Yes	The name of the column that contains dependent variables in the prediction result. The columns of numeric data types are supported.	N/A
indexOutputTableName	Yes	The name of the output table of regression metrics.	N/A
residualOutputTableName	Yes	The name of the output table of the histogram of residuals.	N/A
intervalNum	No	The number of intervals of the histogram.	100
lifecycle	No	The lifecycle of the output table. The value of this parameter must be a positive integer.	N/A
coreNum	No	The number of cores. Valid values: 1 to 9999.	Determined by the system
memSizePerCore	No	The memory size of each core. Valid values: 1024 to 64 × 1024. Unit: MB.	Determined by the system

Output

The output table of regression metrics is generated in the JSON format and contains the following parameters.


Parameter	Description
SST	The total sum of squares.
SSE	The sum of squared errors.
SSR	The sum of squares due to regression.
R2	The coefficient of determination.
R	The coefficient of multiple correlations.
MSE	The mean-square error.
RMSE	The root-mean-square error.
MAE	The mean absolute error.
MAD	The mean error.
MAPE	The mean absolute percentage error.
count	The number of rows.
yMean	The mean of original dependent variables.
predictionMean	The mean of prediction results.