clarena.stl_datasets.kannadamnist

  1r"""
  2The submodule in `stl_datasets` for Kannada-MNIST dataset.
  3"""
  4
  5__all__ = ["KannadaMNIST"]
  6
  7import logging
  8from typing import Callable
  9
 10import torch
 11from torch.utils.data import Dataset, random_split
 12from torchvision.transforms import transforms
 13
 14from clarena.stl_datasets.base import STLDatasetFromRaw
 15from clarena.stl_datasets.raw import KannadaMNIST as KannadaMNISTRaw
 16
 17# always get logger for built-in logging in each module
 18pylogger = logging.getLogger(__name__)
 19
 20
 21class KannadaMNIST(STLDatasetFromRaw):
 22    r"""Kannada-MNIST dataset. The [Kannada-MNIST dataset](https://github.com/vinayprabhu/Kannada_MNIST) is a collection of handwritten Kannada digits (0-9). It consists of 60,000 training and 10,000 test images of handwritten Kannada digits (10 classes), each 28x28 grayscale image (similar to MNIST)."""
 23
 24    original_dataset_python_class: type[Dataset] = KannadaMNISTRaw
 25    r"""The original dataset class."""
 26
 27    def __init__(
 28        self,
 29        root: str,
 30        validation_percentage: float,
 31        batch_size: int = 1,
 32        num_workers: int = 0,
 33        custom_transforms: Callable | transforms.Compose | None = None,
 34        repeat_channels: int | None = None,
 35        to_tensor: bool = True,
 36        resize: tuple[int, int] | None = None,
 37    ) -> None:
 38        r"""
 39        **Args:**
 40        - **root** (`str`): the root directory where the original Kannada-MNIST data 'KannadaMNIST/' live.
 41        - **validation_percentage** (`float`): the percentage to randomly split some training data into validation data.
 42        - **batch_size** (`int`): The batch size in train, val, test dataloader.
 43        - **num_workers** (`int`): the number of workers for dataloaders.
 44        - **custom_transforms** (`transform` or `transforms.Compose` or `None`): the custom transforms to apply to ONLY TRAIN dataset. Can be a single transform, composed transforms or no transform. `ToTensor()`, normalize and so on are not included.
 45        - **repeat_channels** (`int` | `None`): the number of channels to repeat. Default is None, which means no repeat. If not None, it should be an integer.
 46        - **to_tensor** (`bool`): whether to include `ToTensor()` transform. Default is True.
 47        - **resize** (`tuple[int, int]` | `None` or list of them): the size to resize the images to. Default is None, which means no resize. If not None, it should be a tuple of two integers.
 48        """
 49        super().__init__(
 50            root=root,
 51            batch_size=batch_size,
 52            num_workers=num_workers,
 53            custom_transforms=custom_transforms,
 54            repeat_channels=repeat_channels,
 55            to_tensor=to_tensor,
 56            resize=resize,
 57        )
 58
 59        self.validation_percentage: float = validation_percentage
 60        r"""The percentage to randomly split some training data into validation data."""
 61
 62    def prepare_data(self) -> None:
 63        r"""Download the original Kannada-MNIST dataset if haven't. Because the original dataset is published on Kaggle, we need to download it manually. This function will not download the original dataset automatically."""
 64        pass
 65
 66    def train_and_val_dataset(self) -> tuple[Dataset, Dataset]:
 67        """Get the training and validation dataset.
 68
 69        **Returns:**
 70        - **train_and_val_dataset** (`tuple[Dataset, Dataset]`): the train and validation dataset.
 71        """
 72        dataset_train_and_val = KannadaMNISTRaw(
 73            root=self.root,
 74            train=True,
 75            transform=self.train_and_val_transforms(),
 76            target_transform=self.target_transform(),
 77            download=False,
 78        )
 79
 80        return random_split(
 81            dataset_train_and_val,
 82            lengths=[1 - self.validation_percentage, self.validation_percentage],
 83            generator=torch.Generator().manual_seed(
 84                42
 85            ),  # this must be set fixed to make sure the datasets across experiments are the same. Don't handle it to global seed as it might vary across experiments
 86        )
 87
 88    def test_dataset(self) -> Dataset:
 89        r"""Get the test dataset.
 90
 91        **Returns:**
 92        - **test_dataset** (`Dataset`): the test dataset.
 93        """
 94        dataset_test = KannadaMNISTRaw(
 95            root=self.root,
 96            train=False,
 97            transform=self.test_transforms(),
 98            target_transform=self.target_transform(),
 99            download=False,
100        )
101
102        return dataset_test