Minimal Dataset Collection builder.

Users will need to overwrite this function when implementing their dataset collection.

The returned dictionary needs to contain the dataset collection versions as keys, and a dictionary of the included TFDS datasets as values.

@property
def datasets(self):
return {
    "1.0.0":
        naming.references_for({
            "yes_no": "yes_no:1.0.0",
            "sst2": "glue/sst:2.0.0",
            "assin2": "assin2:1.0.0",
        }),
    ...
}

Note that the above is equivalent to:

@property
def datasets(self):
return {
    "1.0.0": {
        "yes_no":
            naming.DatasetReference(
                dataset_name="yes_no", version="1.0.0"),
        "sst2":
            naming.DatasetReference(
                dataset_name="glue", config="sst2", version="2.0.0"),
        "assin2":
            naming.DatasetReference(
                dataset_name="assin2", version="1.0.0"),
    },
    ...
}

Methods

get_collection

View source

get_collection(
    version: Union[None, str, tfds.core.Version] = None
) -> Mapping[str, naming.DatasetReference]

Returns the requested versioned dataset collection.

get_latest_version

View source

get_latest_version() -> str

Returns the latest version of this dataset collection.

list_datasets

View source

list_datasets(
    version: Union[None, str, tfds.core.Version] = None
) -> str

Returns the datasets included in a versioned dataset collection.