HDFDataset

HDFDataset loads and saves data to/from HDF files using pandas.

kedro_datasets.pandas.HDFDataset

HDFDataset(
    *,
    filepath,
    key,
    load_args=None,
    save_args=None,
    version=None,
    credentials=None,
    fs_args=None,
    metadata=None
)

Bases: AbstractVersionedDataset[DataFrame, DataFrame]

HDFDataset loads/saves data from/to a hdf file using an underlying filesystem (e.g. local, S3, GCS). It uses pandas.HDFStore to handle the hdf file.

Examples:

Using the YAML API:

hdf_dataset:
  type: pandas.HDFDataset
  filepath: s3://my_bucket/raw/sensor_reading.h5
  credentials: aws_s3_creds
  key: data

Using the Python API:

>>> import pandas as pd
>>> from kedro_datasets.pandas import HDFDataset
>>>
>>> data = pd.DataFrame({"col1": [1, 2], "col2": [4, 5], "col3": [5, 6]})
>>>
>>> dataset = HDFDataset(filepath=tmp_path / "test.h5", key="data")
>>> dataset.save(data)
>>> reloaded = dataset.load()
>>> assert data.equals(reloaded)

Parameters:

• filepath (str | PathLike) –

  Filepath in POSIX format to a hdf file prefixed with a protocol like s3://. If prefix is not provided, file protocol (local filesystem) will be used. The prefix should be any protocol supported by fsspec. Note: http(s) doesn't support versioning. Can be a string or a PathLike object.

• key (str) –

  Identifier to the group in the HDF store.

• load_args (dict[str, Any] | None, default: None ) –

  PyTables options for loading hdf files. You can find all available arguments at: https://www.pytables.org/usersguide/libref/top_level.html#tables.open_file All defaults are preserved.

• save_args (dict[str, Any] | None, default: None ) –

  PyTables options for saving hdf files. You can find all available arguments at: https://www.pytables.org/usersguide/libref/top_level.html#tables.open_file All defaults are preserved.

• version (Version | None, default: None ) –

  If specified, should be an instance of kedro.io.core.Version. If its load attribute is None, the latest version will be loaded. If its save attribute is None, save version will be autogenerated.

• credentials (dict[str, Any] | None, default: None ) –

  Credentials required to get access to the underlying filesystem. E.g. for GCSFileSystem it should look like {"token": None}.

• fs_args (dict[str, Any] | None, default: None ) –

  Extra arguments to pass into underlying filesystem class constructor (e.g. {"project": "my-project"} for GCSFileSystem), as well as to pass to the filesystem's open method through nested keys open_args_load and open_args_save. Here you can find all available arguments for open: https://filesystem-spec.readthedocs.io/en/latest/api.html#fsspec.spec.AbstractFileSystem.open All defaults are preserved, except open_args_save mode, which is set wb when saving.

• metadata (dict[str, Any] | None, default: None ) –

  Any arbitrary metadata. This is ignored by Kedro, but may be consumed by users or external plugins.

Source code in kedro_datasets/pandas/hdf_dataset.py

def __init__(  # noqa: PLR0913
    self,
    *,
    filepath: str | os.PathLike,
    key: str,
    load_args: dict[str, Any] | None = None,
    save_args: dict[str, Any] | None = None,
    version: Version | None = None,
    credentials: dict[str, Any] | None = None,
    fs_args: dict[str, Any] | None = None,
    metadata: dict[str, Any] | None = None,
) -> None:
    """Creates a new instance of ``HDFDataset`` pointing to a concrete hdf file
    on a specific filesystem.

    Args:
        filepath: Filepath in POSIX format to a hdf file prefixed with a protocol like `s3://`.
            If prefix is not provided, `file` protocol (local filesystem) will be used.
            The prefix should be any protocol supported by ``fsspec``.
            Note: `http(s)` doesn't support versioning.
            Can be a string or a PathLike object.
        key: Identifier to the group in the HDF store.
        load_args: PyTables options for loading hdf files.
            You can find all available arguments at:
            https://www.pytables.org/usersguide/libref/top_level.html#tables.open_file
            All defaults are preserved.
        save_args: PyTables options for saving hdf files.
            You can find all available arguments at:
            https://www.pytables.org/usersguide/libref/top_level.html#tables.open_file
            All defaults are preserved.
        version: If specified, should be an instance of
            ``kedro.io.core.Version``. If its ``load`` attribute is
            None, the latest version will be loaded. If its ``save``
            attribute is None, save version will be autogenerated.
        credentials: Credentials required to get access to the underlying filesystem.
            E.g. for ``GCSFileSystem`` it should look like `{"token": None}`.
        fs_args: Extra arguments to pass into underlying filesystem class constructor
            (e.g. `{"project": "my-project"}` for ``GCSFileSystem``), as well as
            to pass to the filesystem's `open` method through nested keys
            `open_args_load` and `open_args_save`.
            Here you can find all available arguments for `open`:
            https://filesystem-spec.readthedocs.io/en/latest/api.html#fsspec.spec.AbstractFileSystem.open
            All defaults are preserved, except `open_args_save` `mode`, which is set `wb` when saving.
        metadata: Any arbitrary metadata.
            This is ignored by Kedro, but may be consumed by users or external plugins.
    """
    _fs_args = deepcopy(fs_args) or {}
    _fs_open_args_load = _fs_args.pop("open_args_load", {})
    _fs_open_args_save = _fs_args.pop("open_args_save", {})
    _credentials = deepcopy(credentials) or {}

    protocol, path = get_protocol_and_path(filepath, version)
    if protocol == "file":
        _fs_args.setdefault("auto_mkdir", True)

    self._protocol = protocol
    self._fs = fsspec.filesystem(self._protocol, **_credentials, **_fs_args)

    self.metadata = metadata

    super().__init__(
        filepath=PurePosixPath(path),
        version=version,
        exists_function=self._fs.exists,
        glob_function=self._fs.glob,
    )

    self._key = key

    # Handle default load and save and fs arguments
    self._load_args = {**self.DEFAULT_LOAD_ARGS, **(load_args or {})}
    self._save_args = {**self.DEFAULT_SAVE_ARGS, **(save_args or {})}
    self._fs_open_args_load = {
        **self.DEFAULT_FS_ARGS.get("open_args_load", {}),
        **(_fs_open_args_load or {}),
    }
    self._fs_open_args_save = {
        **self.DEFAULT_FS_ARGS.get("open_args_save", {}),
        **(_fs_open_args_save or {}),
    }

DEFAULT_FS_ARGS class-attribute instance-attribute

DEFAULT_FS_ARGS = {'open_args_save': {'mode': 'wb'}}

DEFAULT_LOAD_ARGS class-attribute instance-attribute

DEFAULT_LOAD_ARGS = {}

DEFAULT_SAVE_ARGS class-attribute instance-attribute

DEFAULT_SAVE_ARGS = {}

_fs instance-attribute

_fs = filesystem(_protocol, **_credentials, **_fs_args)

_fs_open_args_load instance-attribute

_fs_open_args_load = {
    None: get("open_args_load", {}),
    None: _fs_open_args_load or {},
}

_fs_open_args_save instance-attribute

_fs_open_args_save = {
    None: get("open_args_save", {}),
    None: _fs_open_args_save or {},
}

_key instance-attribute

_key = key

_load_args instance-attribute

_load_args = {
    None: DEFAULT_LOAD_ARGS,
    None: load_args or {},
}

_lock class-attribute instance-attribute

_lock = Lock()

_protocol instance-attribute

_protocol = protocol

_save_args instance-attribute

_save_args = {
    None: DEFAULT_SAVE_ARGS,
    None: save_args or {},
}

metadata instance-attribute

metadata = metadata

_describe

_describe()

Source code in kedro_datasets/pandas/hdf_dataset.py

def _describe(self) -> dict[str, Any]:
    return {
        "filepath": self._filepath,
        "key": self._key,
        "protocol": self._protocol,
        "load_args": self._load_args,
        "save_args": self._save_args,
        "version": self._version,
    }

_exists

_exists()

Source code in kedro_datasets/pandas/hdf_dataset.py

def _exists(self) -> bool:
    try:
        load_path = get_filepath_str(self._get_load_path(), self._protocol)
    except DatasetError:
        return False

    return self._fs.exists(load_path)

_invalidate_cache

_invalidate_cache()

Invalidate underlying filesystem caches.

Source code in kedro_datasets/pandas/hdf_dataset.py

def _invalidate_cache(self) -> None:
    """Invalidate underlying filesystem caches."""
    filepath = get_filepath_str(self._filepath, self._protocol)
    self._fs.invalidate_cache(filepath)

_release

_release()

Source code in kedro_datasets/pandas/hdf_dataset.py

def _release(self) -> None:
    super()._release()
    self._invalidate_cache()

load

load()

Source code in kedro_datasets/pandas/hdf_dataset.py

def load(self) -> pd.DataFrame:
    load_path = get_filepath_str(self._get_load_path(), self._protocol)

    with self._fs.open(load_path, **self._fs_open_args_load) as fs_file:
        binary_data = fs_file.read()

    with HDFDataset._lock:
        # Set driver_core_backing_store to False to disable saving
        # contents of the in-memory h5file to disk
        with pd.HDFStore(
            "in-memory-load-file",
            mode="r",
            driver=HDFSTORE_DRIVER,
            driver_core_backing_store=0,
            driver_core_image=binary_data,
            **self._load_args,
        ) as store:
            return store[self._key]

save

save(data)

Source code in kedro_datasets/pandas/hdf_dataset.py

def save(self, data: pd.DataFrame) -> None:
    save_path = get_filepath_str(self._get_save_path(), self._protocol)

    with HDFDataset._lock:
        with pd.HDFStore(
            "in-memory-save-file",
            mode="w",
            driver=HDFSTORE_DRIVER,
            driver_core_backing_store=0,
            **self._save_args,
        ) as store:
            store.put(self._key, data, format="table")
            binary_data = store._handle.get_file_image()

    with self._fs.open(save_path, **self._fs_open_args_save) as fs_file:
        fs_file.write(binary_data)

    self._invalidate_cache()