Linear regression is a statistical method that models the linear relationship between a dependent variable and one or more independent variables. It fits an optimal linear curve to the data, minimizing the error between predicted and actual values. This process lets you estimate and analyze the dependent variable.
Configure the component
Method 1: Use the UI
Add the Linear Regression component to the workflow page in Designer, and configure its parameters in the pane on the right:
|
Parameter type |
Parameter |
Description |
|
Fields setting |
Select feature columns |
The feature columns from the input data source to use for training. |
|
Select label column |
The DOUBLE and BIGINT types are supported. |
|
|
Is sparse format |
You can represent the sparse format in the KV format. |
|
|
Separator between key-value pairs |
The default separator is a comma (,). |
|
|
Separator between key and value |
The separator between keys and values. The default is a colon (:). |
|
|
Parameters setting |
Maximum iterations |
The maximum number of iterations for the algorithm. |
|
Minimum likelihood error |
Uses log-likelihood as the convergence criterion. The algorithm stops if the difference in log-likelihood between two consecutive iterations is less than this value. |
|
|
Regularization type |
The regularization type. Supported types are L1, L2, and None. |
|
|
Regularization coefficient |
If Regularization Type is set to None, this parameter is ignored. |
|
|
Generate model evaluation table |
The metrics include R-Squared, Adjusted R-Squared, AIC, degrees of freedom, the standard deviation of the residuals, and bias. |
|
|
Regression coefficient evaluation |
Metrics include T-value, P-value, and the confidence interval [2.5%,97.5%]. This parameter is available only if the Generate model evaluation table check box is selected. |
|
|
Execution tuning |
Number of computing cores |
By default, the system automatically allocates the cores. |
|
Memory size per core |
By default, the system automatically allocates the memory. |
Method 2: Use PAI commands
Use PAI commands to configure the parameters for the Linear Regression component. You can use the SQL script component to call PAI commands. For more information, see SQL Script.
PAI -name linearregression
-project algo_public
-DinputTableName=lm_test_input
-DfeatureColNames=x
-DlabelColName=y
-DmodelName=lm_test_input_model_out;|
Parameter |
Required |
Default value |
Description |
|
inputTableName |
Yes |
None |
The name of the input table. |
|
modelName |
Yes |
None |
The name of the output model. |
|
outputTableName |
No |
None |
The name of the output model evaluation table. This parameter is required if enableFitGoodness is true. |
|
labelColName |
Yes |
None |
The dependent variable. DOUBLE and BIGINT types are supported. You can select only one column as the dependent variable. |
|
featureColNames |
Yes |
None |
The independent variables. If the input data is in dense format, DOUBLE and BIGINT types are supported. If the input data is in sparse format, the STRING type is supported. |
|
inputTablePartitions |
No |
None |
The partitions of the input table. |
|
enableSparse |
No |
false |
Specifies whether the input data is in sparse format. Valid values are {true,false}. |
|
itemDelimiter |
No |
Comma (,) |
The separator between key-value pairs. This parameter is used if enableSparse is true. |
|
kvDelimiter |
No |
Colon (:) |
The separator between a key and a value. This parameter is used if enableSparse is true. |
|
maxIter |
No |
100 |
The maximum number of iterations for the algorithm. |
|
epsilon |
No |
0.000001 |
The minimum likelihood error. The algorithm stops if the difference in log-likelihood between two consecutive iterations is less than this value. |
|
regularizedType |
No |
None |
The regularization type. Valid values are {l1,l2,None}. |
|
regularizedLevel |
No |
1 |
The regularization coefficient. This parameter is not used if regularizedType is None. |
|
enableFitGoodness |
No |
false |
Specifies whether to generate a model evaluation table. Metrics include R-Squared, Adjusted R-Squared, AIC, degrees of freedom, standard deviation of residuals, and deviance. Valid values are {true,false}. |
|
enableCoefficientEstimate |
No |
false |
Specifies whether to evaluate the regression coefficient. Evaluation metrics include T-value, P-value, and the confidence interval [2.5%,97.5%]. This parameter is used if enableFitGoodness is true. Valid values are {true,false}. |
|
lifecycle |
No |
-1 |
The lifecycle of the output model evaluation table. |
|
coreNum |
No |
System allocated |
By default, the system automatically allocates the cores. |
|
memSizePerCore |
No |
System allocated |
By default, the system automatically allocates the memory. |
Examples
-
Generate test data using an SQL statement.
drop table if exists lm_test_input; create table lm_test_input as select * from ( select 10 as y, 1.84 as x1, 1 as x2, '0:1.84 1:1' as sparsecol1 union all select 20 as y, 2.13 as x1, 0 as x2, '0:2.13' as sparsecol1 union all select 30 as y, 3.89 as x1, 0 as x2, '0:3.89' as sparsecol1 union all select 40 as y, 4.19 as x1, 0 as x2, '0:4.19' as sparsecol1 union all select 50 as y, 5.76 as x1, 0 as x2, '0:5.76' as sparsecol1 union all select 60 as y, 6.68 as x1, 2 as x2, '0:6.68 1:2' as sparsecol1 union all select 70 as y, 7.58 as x1, 0 as x2, '0:7.58' as sparsecol1 union all select 80 as y, 8.01 as x1, 0 as x2, '0:8.01' as sparsecol1 union all select 90 as y, 9.02 as x1, 3 as x2, '0:9.02 1:3' as sparsecol1 union all select 100 as y, 10.56 as x1, 0 as x2, '0:10.56' as sparsecol1 ) tmp; -
Submit the parameters for the linear regression component using a PAI command.
PAI -name linearregression -project algo_public -DinputTableName=lm_test_input -DlabelColName=y -DfeatureColNames=x1,x2 -DmodelName=lm_test_input_model_out -DoutputTableName=lm_test_input_conf_out -DenableCoefficientEstimate=true -DenableFitGoodness=true -Dlifecycle=1; -
Submit the parameters for the prediction component using a PAI command.
pai -name prediction -project algo_public -DmodelName=lm_test_input_model_out -DinputTableName=lm_test_input -DoutputTableName=lm_test_input_predict_out -DappendColNames=y; -
View the output model evaluation table lm_test_input_conf_out.
+------------+------------+------------+------------+--------------------+------------+ | colname | value | tscore | pvalue | confidenceinterval | p | +------------+------------+------------+------------+--------------------+------------+ | Intercept | -6.42378496687763 | -2.2725755951390028 | 0.06 | {"2.5%": -11.964027, "97.5%": -0.883543} | coefficient | | x1 | 10.260063429838898 | 23.270944360826963 | 0.0 | {"2.5%": 9.395908, "97.5%": 11.124219} | coefficient | | x2 | 0.35374498323846265 | 0.2949247320997519 | 0.81 | {"2.5%": -1.997160, "97.5%": 2.704650} | coefficient | | rsquared | 0.9879675667384592 | NULL | NULL | NULL | goodness | | adjusted_rsquared | 0.9845297286637332 | NULL | NULL | NULL | goodness | | aic | 59.331109494251805 | NULL | NULL | NULL | goodness | | degree_of_freedom | 7.0 | NULL | NULL | NULL | goodness | | standardErr_residual | 3.765777749448906 | NULL | NULL | NULL | goodness | | deviance | 99.26757440771128 | NULL | NULL | NULL | goodness | +------------+------------+------------+------------+--------------------+------------+ -
View the prediction result table lm_test_input_predict_out.
+------------+-------------------+------------------+-------------------+ | y | prediction_result | prediction_score | prediction_detail | +------------+-------------------+------------------+-------------------+ | 10 | NULL | 12.808476727264404 | {"y": 12.8084767272644} | | 20 | NULL | 15.43015013867922 | {"y": 15.43015013867922} | | 30 | NULL | 33.48786177519568 | {"y": 33.48786177519568} | | 40 | NULL | 36.565880804147355 | {"y": 36.56588080414735} | | 50 | NULL | 52.674180388994415 | {"y": 52.67418038899442} | | 60 | NULL | 62.82092871092313 | {"y": 62.82092871092313} | | 70 | NULL | 71.34749583130122 | {"y": 71.34749583130122} | | 80 | NULL | 75.75932310613193 | {"y": 75.75932310613193} | | 90 | NULL | 87.1832221199846 | {"y": 87.18322211998461} | | 100 | NULL | 101.92248485222113 | {"y": 101.9224848522211} | +------------+-------------------+------------------+-------------------+