The SparseSVMPredictor function has two required input tables:
- The sample table, which contains the test data
The following table describes the required sample table columns. The function ignores any additional columns, except those specified by the AccumulateLabel argument, which it copies to the output table.
- The model table, which is output by the SparseSVMTrainer function
The model table is in binary format. To display its readable content, use the function SVMModelPrinter.
Column Name | Data Type | Description |
---|---|---|
id_column | INTEGER, SMALLINT, BIGINT, NUMERIC, NUMERIC(p), NUMERIC(p,a), TEXT, VARCHAR, VARCHAR(n), UUID, or BYTEA | Contains the identifiers of the test samples. |
attribute_column | INTEGER, SMALLINT, BIGINT, TEXT, VARCHAR, or VARCHAR(n) | Contains the attributes of the test samples. |
value_column | INTEGER, SMALLINT, BIGINT, NUMERIC, NUMERIC(p), or NUMERIC(p,a) | Contains the attribute values. |