Skip to content

Datasets

Ready-to-use datasets.

A set of toy datasets formated for somap Self-Organizing Maps.

Digits

Bases: SomDataset

Kind of mini-MNIST.

~2k grayscale 8x8 images of digits from 0 to 9.

Source code in src/somap/datasets/mnist.py 
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
class Digits(SomDataset):
    """Kind of mini-MNIST.

    ~2k grayscale 8x8 images of digits from 0 to 9.
    """

    def __init__(self):
        """Loads the mini-MNIST dataset from scikit-learn."""

        def _load_func():
            dataset = load_dataset("sklearn-docs/digits", header=None)
            return np.asarray(dataset["train"].data).T[:, :64].reshape(-1, 8, 8) / 16.0

        self.data = load_data_with_caching("digits.npy", _load_func)
        self.nb = self.data.shape[0]
        self.shape = self.data.shape[1:]

__init__()

Loads the mini-MNIST dataset from scikit-learn.

Source code in src/somap/datasets/mnist.py 
34
35
36
37
38
39
40
41
42
43
def __init__(self):
    """Loads the mini-MNIST dataset from scikit-learn."""

    def _load_func():
        dataset = load_dataset("sklearn-docs/digits", header=None)
        return np.asarray(dataset["train"].data).T[:, :64].reshape(-1, 8, 8) / 16.0

    self.data = load_data_with_caching("digits.npy", _load_func)
    self.nb = self.data.shape[0]
    self.shape = self.data.shape[1:]

MNIST

Bases: SomDataset

Classic MNIST.

60k grayscale 28x28 images of digits from 0 to 9.

Source code in src/somap/datasets/mnist.py 
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
class MNIST(SomDataset):
    """Classic MNIST.

    60k grayscale 28x28 images of digits from 0 to 9.
    """

    def __init__(self):
        """Loads the classic MNIST dataset."""

        def _load_func():
            dataset = load_dataset("mnist")
            return np.asarray(dataset["train"]["image"]) / 255.0

        self.data = load_data_with_caching("mnist.npy", _load_func)
        self.nb = self.data.shape[0]
        self.shape = self.data.shape[1:]

__init__()

Loads the classic MNIST dataset.

Source code in src/somap/datasets/mnist.py 
16
17
18
19
20
21
22
23
24
25
def __init__(self):
    """Loads the classic MNIST dataset."""

    def _load_func():
        dataset = load_dataset("mnist")
        return np.asarray(dataset["train"]["image"]) / 255.0

    self.data = load_data_with_caching("mnist.npy", _load_func)
    self.nb = self.data.shape[0]
    self.shape = self.data.shape[1:]