Spark Numeric Scorer

This node computes certain statistics between the a numeric column's values (r_i) and predicted (p_i) values. It computes R²=1-SS_res/SS_tot=1-Σ(p_i-r_i)²/Σ(r_i-1/n*Σr_i)² (can be negative!), mean absolute error (1/n*Σ|p_i-r_i|), mean squared error (1/n*Σ(p_i-r_i)²), root mean squared error (sqrt(1/n*Σ(p_i-r_i)²)), and mean signed difference (1/n*Σ(p_i-r_i)). The computed values can be inspected in the node's view and/or further processed using the output table.

Options

Reference column

Column with the correct, observed, training data values. Rows with missing values in selected column will be ignored.

Predicted column

Column with the modeled, predicted data values. Computation will fail if selected column contains missing values.

Change column name

Change the default output column name.

Output column name

The name of the column in the output.

Output scores as flow variables

The scores can be exported as flow variables.

Prefix of flow variables

This option allows you to define a prefix for these variable identifiers so that name conflicts are resolved.

Views

Statistics

A table with the statistical measures

Workflows

• 01 Taxi Demand Prediction on Spark TrainingKNIME Hub
• 02_Taxi_Demand_Prediction_Training_workflowKNIME Hub
• 02_Taxi_Demand_Prediction_Training_workflowKNIME Hub
• 08_Connecting_to_Amazon_EMRKNIME Hub
• 10_Recommendation_Engine_w_Spark_Collaborative_FilteringKNIME Hub
• Show all 14 workflows

Links

• No links available

Developers