doubt.datasets.airfoil

 1"""Airfoil 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 Airfoil(BaseDataset):
16    __doc__ = f"""
17    The NASA data set comprises different size NACA 0012 airfoils at various wind
18    tunnel speeds and angles of attack. The span of the airfoil and the observer
19    position were the same in all of the experiments.
20
21    {BASE_DATASET_DESCRIPTION}
22
23    Features:
24        int:
25            Frequency, in Hertzs
26        float:
27            Angle of attack, in degrees
28        float:
29            Chord length, in meters
30        float:
31            Free-stream velocity, in meters per second
32        float:
33            Suction side displacement thickness, in meters
34
35    Targets:
36        float:
37            Scaled sound pressure level, in decibels
38
39    Source:
40        https://archive.ics.uci.edu/ml/datasets/Airfoil+Self-Noise
41
42    Examples:
43        Load in the data set::
44
45            >>> dataset = Airfoil()
46            >>> dataset.shape
47            (1503, 6)
48
49        Split the data set into features and targets, as NumPy arrays::
50
51            >>> X, y = dataset.split()
52            >>> X.shape, y.shape
53            ((1503, 5), (1503,))
54
55        Perform a train/test split, also outputting NumPy arrays::
56
57            >>> train_test_split = dataset.split(test_size=0.2, random_seed=42)
58            >>> X_train, X_test, y_train, y_test = train_test_split
59            >>> X_train.shape, y_train.shape, X_test.shape, y_test.shape
60            ((1181, 5), (1181,), (322, 5), (322,))
61
62        Output the underlying Pandas DataFrame::
63
64            >>> df = dataset.to_pandas()
65            >>> type(df)
66            <class 'pandas.core.frame.DataFrame'>
67    """
68
69    _url = (
70        "https://archive.ics.uci.edu/ml/machine-learning-databases/"
71        "00291/airfoil_self_noise.dat"
72    )
73
74    _features = range(5)
75    _targets = [5]
76
77    def _prep_data(self, data: bytes) -> pd.DataFrame:
78        """Prepare the data set.
79
80        Args:
81            data (bytes): The raw data
82
83        Returns:
84            Pandas dataframe: The prepared data
85        """
86        # Convert the bytes into a file-like object
87        csv_file = io.BytesIO(data)
88
89        # Read the file-like object into a data frame
90        df = pd.read_csv(csv_file, sep="\t", header=None)
91        return df