Merged Dataset

The merged dataset allows for rows from multiple datasets to be merged into a single dataset. Rows and columns that match up between the datasets will be combined together.

The merge is done on the fly (only an index is created of rows and columns in the merged dataset), which means it is relatively rapid to merge even large datasets together. However, this means the underlying datasets that are being merged cannot be deleted.

Creating a merged dataset is equivalent to the following SQL:

SELECT * FROM ds1 OUTER JOIN ds2 ON ds1.rowName() = ds2.rowName()

Configuration

A new dataset of type merged named <id> can be created as follows:

mldb.put("/v1/datasets/"+<id>, {
    "type": "merged",
    "params": {
        "datasets": <ARRAY [ Dataset (read-only) ]>
    }
})

with the following key-value definitions for params:

See Also

• The merge function can also be used within From expressions.
• The transposed dataset type is another dataset transformation
• The transform procedure type can be used to modify a dataset ready for merging