Column | Data Type | Description |
---|---|---|
accumulate_column | Same as in InputTable. | One column each for the column names specified in accumulate_column. Use an asterisk to specify all columns. |
prediction | VARCHAR | The prediction value. Appears when the JSON has only one entry. Often occurs during regression for both H2O Open Source and DAI models. If a classification, the probabilities of the classes are included in the JSON report. |
json_report | VARCHAR | All output fields specific to the model, except for prediction. If EnableOptions is used, those fields are included. Appears only if you omit ModelOutputFields. |
model_output_field | VARCHAR or DOUBLE PRECISION | The model-specific output fields such as probability, predicted, and value. Appears once for each specified model_output_field. If EnableOptions is set, an extra field for the option is added to the JSON report. If any of the options are to be a separate column outside of the JSON, the option needs to also be included in the ModelOutputFields parameter. |