my_project.dataset

Dataset utilities for the automobile price prediction model.

 1"""Dataset utilities for the automobile price prediction model."""
 2
 3import pandas as pd
 4import torch
 5from torch.utils.data import Dataset
 6
 7
 8class AutomobileDataset(Dataset):
 9    """PyTorch dataset for automobile price prediction.
10
11    Loads a processed Parquet dataset, separates the input features from
12    the target variable, and converts both to PyTorch tensors.
13
14    Parameters
15    ----------
16    path : str
17        Path to the Parquet file containing the processed automobile data.
18
19    Attributes
20    ----------
21    X : torch.Tensor
22        Input features as a two-dimensional float32 tensor.
23    y : torch.Tensor
24        Target selling prices as a one-dimensional float32 tensor.
25
26    Examples
27    --------
28    >>> dataset = AutomobileDataset("data/processed/automobile_dataset.parquet")
29    >>> len(dataset)
30    # Number of samples in the dataset
31    """
32
33    def __init__(self, path: str):
34        """Initialize the automobile dataset.
35
36        Parameters
37        ----------
38        path : str
39            Path to the Parquet dataset.
40        """
41        df = pd.read_parquet(path)
42
43        self.X = torch.tensor(
44            df.drop(columns=["Selling_Price"]).values,
45            dtype=torch.float32
46        )
47
48        self.y = torch.tensor(
49            df["Selling_Price"].values,
50            dtype=torch.float32
51        )
52
53    def __len__(self) -> int:
54        """Return the number of samples in the dataset.
55
56        Returns
57        -------
58        int
59            Number of samples available in the dataset.
60        """
61        return len(self.X)
62
63    def __getitem__(self, index: int) -> tuple[torch.Tensor, torch.Tensor]:
64        """Return a single sample from the dataset.
65
66        Parameters
67        ----------
68        index : int
69            Index of the sample to retrieve.
70
71        Returns
72        -------
73        tuple[torch.Tensor, torch.Tensor]
74            A tuple containing the input features and target selling price.
75        """
76        return self.X[index], self.y[index]
class AutomobileDataset(typing.Generic[+_T_co]): 
 9class AutomobileDataset(Dataset):
10    """PyTorch dataset for automobile price prediction.
11
12    Loads a processed Parquet dataset, separates the input features from
13    the target variable, and converts both to PyTorch tensors.
14
15    Parameters
16    ----------
17    path : str
18        Path to the Parquet file containing the processed automobile data.
19
20    Attributes
21    ----------
22    X : torch.Tensor
23        Input features as a two-dimensional float32 tensor.
24    y : torch.Tensor
25        Target selling prices as a one-dimensional float32 tensor.
26
27    Examples
28    --------
29    >>> dataset = AutomobileDataset("data/processed/automobile_dataset.parquet")
30    >>> len(dataset)
31    # Number of samples in the dataset
32    """
33
34    def __init__(self, path: str):
35        """Initialize the automobile dataset.
36
37        Parameters
38        ----------
39        path : str
40            Path to the Parquet dataset.
41        """
42        df = pd.read_parquet(path)
43
44        self.X = torch.tensor(
45            df.drop(columns=["Selling_Price"]).values,
46            dtype=torch.float32
47        )
48
49        self.y = torch.tensor(
50            df["Selling_Price"].values,
51            dtype=torch.float32
52        )
53
54    def __len__(self) -> int:
55        """Return the number of samples in the dataset.
56
57        Returns
58        -------
59        int
60            Number of samples available in the dataset.
61        """
62        return len(self.X)
63
64    def __getitem__(self, index: int) -> tuple[torch.Tensor, torch.Tensor]:
65        """Return a single sample from the dataset.
66
67        Parameters
68        ----------
69        index : int
70            Index of the sample to retrieve.
71
72        Returns
73        -------
74        tuple[torch.Tensor, torch.Tensor]
75            A tuple containing the input features and target selling price.
76        """
77        return self.X[index], self.y[index]

PyTorch dataset for automobile price prediction.

Loads a processed Parquet dataset, separates the input features from the target variable, and converts both to PyTorch tensors.

Parameters
  • path (str): Path to the Parquet file containing the processed automobile data.
Attributes
  • X (torch.Tensor): Input features as a two-dimensional float32 tensor.
  • y (torch.Tensor): Target selling prices as a one-dimensional float32 tensor.
Examples
>>> dataset = AutomobileDataset("data/processed/automobile_dataset.parquet")
>>> len(dataset)
<h1 id="number-of-samples-in-the-dataset">Number of samples in the dataset</h1>
AutomobileDataset(path: str) 
34    def __init__(self, path: str):
35        """Initialize the automobile dataset.
36
37        Parameters
38        ----------
39        path : str
40            Path to the Parquet dataset.
41        """
42        df = pd.read_parquet(path)
43
44        self.X = torch.tensor(
45            df.drop(columns=["Selling_Price"]).values,
46            dtype=torch.float32
47        )
48
49        self.y = torch.tensor(
50            df["Selling_Price"].values,
51            dtype=torch.float32
52        )

Initialize the automobile dataset.

Parameters
  • path (str): Path to the Parquet dataset.
X
y