GetCSVStorage returns the instantiated csv storage.

GetParquetStorage returns the instantiated parquet storage.

GetStorage returns the storage to use based on URI.

NewCSV creates a new csv backed storage.

NewParquet creates a new parquet backed storage.

ReadDataset reads the metadata to find the main data reference, then reads that.

ReadMetadata reads the metadata in the specified path.

ResultToInputCSV takes a result produced by a TA2 pipeline run ensures that it is in a format suitable for storage as a D3M dataset.

WriteData writes data to storage using the specified URI.

WriteDataset determines which storage engine to use and then writes out the metadata and the data using it.

WriteMetadata writes the metadata to disk.

CSV represents a dataset storage backed with csv data and json schema doc.

Parquet represents a dataset storage backed with parquet data and json schema doc.

RawDataset contains basic information about the structure of the dataset as well as the raw learning data.

Storage defines the base functions needed to store datasets to a backing storage for interactions with an auto ml server.