doubt.datasets.cpu

  1"""CPU data set.
  2
  3This data set is from the UCI data set archive, with the description being the original
  4description verbatim. Some feature names may have been altered, based on the
  5description.
  6"""
  7
  8import io
  9
 10import pandas as pd
 11
 12from .dataset import BASE_DATASET_DESCRIPTION, BaseDataset
 13
 14
 15class CPU(BaseDataset):
 16    __doc__ = f"""
 17    Relative CPU Performance Data, described in terms of its cycle time, memory size,
 18    etc.
 19
 20    {BASE_DATASET_DESCRIPTION}
 21
 22    Features:
 23        vendor_name (string):
 24            Name of the vendor, 30 unique values
 25        model_name (string):
 26            Name of the model
 27        myct (int):
 28            Machine cycle time in nanoseconds
 29        mmin (int):
 30            Minimum main memory in kilobytes
 31        mmax (int):
 32            Maximum main memory in kilobytes
 33        cach (int):
 34            Cache memory in kilobytes
 35        chmin (int):
 36            Minimum channels in units
 37        chmax (int):
 38            Maximum channels in units
 39
 40    Targets:
 41        prp (int):
 42            Published relative performance
 43
 44    Source:
 45        https://archive.ics.uci.edu/ml/datasets/Computer+Hardware
 46
 47    Examples:
 48        Load in the data set::
 49
 50            >>> dataset = CPU()
 51            >>> dataset.shape
 52            (209, 9)
 53
 54        Split the data set into features and targets, as NumPy arrays::
 55
 56            >>> X, y = dataset.split()
 57            >>> X.shape, y.shape
 58            ((209, 8), (209,))
 59
 60        Perform a train/test split, also outputting NumPy arrays::
 61
 62            >>> train_test_split = dataset.split(test_size=0.2, random_seed=42)
 63            >>> X_train, X_test, y_train, y_test = train_test_split
 64            >>> X_train.shape, y_train.shape, X_test.shape, y_test.shape
 65            ((162, 8), (162,), (47, 8), (47,))
 66
 67        Output the underlying Pandas DataFrame::
 68
 69            >>> df = dataset.to_pandas()
 70            >>> type(df)
 71            <class 'pandas.core.frame.DataFrame'>
 72    """
 73
 74    _url = (
 75        "https://archive.ics.uci.edu/ml/machine-learning-databases/"
 76        "cpu-performance/machine.data"
 77    )
 78
 79    _features = range(8)
 80    _targets = [8]
 81
 82    def _prep_data(self, data: bytes) -> pd.DataFrame:
 83        """Prepare the data set.
 84
 85        Args:
 86            data (bytes): The raw data
 87
 88        Returns:
 89            Pandas dataframe: The prepared data
 90        """
 91
 92        # Convert the bytes into a file-like object
 93        csv_file = io.BytesIO(data)
 94
 95        # Name the columns
 96        cols = [
 97            "vendor_name",
 98            "model_name",
 99            "myct",
100            "mmin",
101            "mmax",
102            "cach",
103            "chmin",
104            "chmax",
105            "prp",
106        ]
107
108        # Load the file-like object into a data frame
109        df = pd.read_csv(csv_file, header=None, usecols=range(9), names=cols)
110        return df