doubt.datasets.new_taipei_housing

  1"""New Taipei Housing 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 NewTaipeiHousing(BaseDataset):
 16    __doc__ = f"""
 17    The "real estate valuation" is a regression problem. The market historical data set
 18    of real estate valuation are collected from Sindian Dist., New Taipei City, Taiwan.
 19
 20    {BASE_DATASET_DESCRIPTION}
 21
 22    Features:
 23        transaction_date (float):
 24            The transaction date encoded as a floating point value. For instance,
 25            2013.250 is March 2013 and 2013.500 is June March
 26        house_age (float):
 27            The age of the house
 28        mrt_distance (float):
 29            Distance to the nearest MRT station
 30        n_stores (int):
 31            Number of convenience stores
 32        lat (float):
 33            Latitude
 34        lng (float):
 35            Longitude
 36
 37    Targets:
 38        house_price (float):
 39            House price of unit area
 40
 41    Source:
 42        https://archive.ics.uci.edu/ml/datasets/Real+estate+valuation+data+set
 43
 44    Examples:
 45        Load in the data set::
 46
 47            >>> dataset = NewTaipeiHousing()
 48            >>> dataset.shape
 49            (414, 7)
 50
 51        Split the data set into features and targets, as NumPy arrays::
 52
 53            >>> X, y = dataset.split()
 54            >>> X.shape, y.shape
 55            ((414, 6), (414,))
 56
 57        Perform a train/test split, also outputting NumPy arrays::
 58
 59            >>> train_test_split = dataset.split(test_size=0.2, random_seed=42)
 60            >>> X_train, X_test, y_train, y_test = train_test_split
 61            >>> X_train.shape, y_train.shape, X_test.shape, y_test.shape
 62            ((323, 6), (323,), (91, 6), (91,))
 63
 64        Output the underlying Pandas DataFrame::
 65
 66            >>> df = dataset.to_pandas()
 67            >>> type(df)
 68            <class 'pandas.core.frame.DataFrame'>
 69    """
 70
 71    _url = (
 72        "https://archive.ics.uci.edu/ml/machine-learning-databases/"
 73        "00477/Real%20estate%20valuation%20data%20set.xlsx"
 74    )
 75
 76    _features = range(6)
 77    _targets = [6]
 78
 79    def _prep_data(self, data: bytes) -> pd.DataFrame:
 80        """Prepare the data set.
 81
 82        Args:
 83            data (bytes): The raw data
 84
 85        Returns:
 86            Pandas dataframe: The prepared data
 87        """
 88        # Convert the bytes into a file-like object
 89        xlsx_file = io.BytesIO(data)
 90
 91        # Load in the dataframe
 92        cols = [
 93            "idx",
 94            "transaction_date",
 95            "house_age",
 96            "mrt_distance",
 97            "n_stores",
 98            "lat",
 99            "lng",
100            "house_price",
101        ]
102        df = pd.read_excel(xlsx_file, header=0, names=cols)
103
104        # Remove the index
105        df = df.iloc[:, 1:]
106
107        return df