cli

---

parse_shape(value)

Parse a shape argument into a tuple of integers.

This function takes a string representing shape dimensions. If the input is a single value, it is duplicated to form a tuple of length two. If multiple values are provided, they are converted into a tuple.

Args: value: A string containing one or more integers separated by commas, representing shape dimensions.

Returns: A tuple of two integers derived from the input string.

Examples: parse_shape("5") returns (5, 5) parse_shape("5,6") returns (5, 6)

Source code in preprocess_toolbox/cli.py

def parse_shape(value: str) -> tuple[int, int]:
    """
    Parse a shape argument into a tuple of integers.

    This function takes a string representing shape dimensions.
    If the input is a single value, it is duplicated to form a tuple of
    length two. If multiple values are provided, they are converted into
    a tuple.

    Args:
        value: A string containing one or more integers separated by commas,
            representing shape dimensions.

    Returns:
        A tuple of two integers derived from the input string.

    Examples:
        parse_shape("5") returns (5, 5)
        parse_shape("5,6") returns (5, 6)
    """
    if isinstance(value, int):
        return (value, value)
    else:
        values = value.split(",")

        # If only one value is provided, repeat it to create a tuple of length 2
        if len(values) == 1:
            values.append(values[0])
        values = map(int, values)

    return tuple(values)

process_split_args(args, frequency)

Parameters:

Name	Type	Description	Default
args	object		required
frequency	Frequency		required

Returns:

Type	Description
dict

Source code in preprocess_toolbox/cli.py

def process_split_args(args: object,
                       frequency: Frequency) -> dict:
    """

    :param args:
    :param frequency:
    :return:
    """
    if not hasattr(args, "split_names") or args.split_names is None:
        logging.info("No split names in arguments to process...")
        return dict()

    splits = {_: list() for _ in args.split_names}

    for idx, split in enumerate(splits.keys()):
        split_dates = collections.deque()

        for period_start, period_end in zip(args.split_starts[idx], args.split_ends[idx]):
            split_dates += [
                pd.to_datetime(date).date()
                for date in pd.date_range(period_start, period_end, freq=frequency.freq)
            ]
        logging.info("Got {} dates for {}".format(len(split_dates), split))

        splits[split] = sorted(list(split_dates))
    return splits