Skip to content

horse2zebra

load_data

Load and return the horse2zebra dataset.

Sourced from https://people.eecs.berkeley.edu/~taesung_park/CycleGAN/datasets/horse2zebra.zip. This method will download the data to local storage if the data has not been previously downloaded.

Parameters:

Name Type Description Default
batch_size int

The desired batch size.

 required
root_dir Optional[str]

The path to store the downloaded data. When path is not provided, the data will be saved into fastestimator_data under the user's home directory.

 None

Returns:

Type Description
Tuple[BatchDataset, BatchDataset]

(train_data, eval_data)
Source code in fastestimator/fastestimator/dataset/data/horse2zebra.py 
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
def load_data(batch_size: int, root_dir: Optional[str] = None) -> Tuple[BatchDataset, BatchDataset]:
    """Load and return the horse2zebra dataset.

    Sourced from https://people.eecs.berkeley.edu/~taesung_park/CycleGAN/datasets/horse2zebra.zip. This method will
        download the data to local storage if the data has not been previously downloaded.

    Args:
        batch_size: The desired batch size.
        root_dir: The path to store the downloaded data. When `path` is not provided, the data will be saved into
            `fastestimator_data` under the user's home directory.

    Returns:
        (train_data, eval_data)
    """
    home = str(Path.home())

    if root_dir is None:
        root_dir = os.path.join(home, 'fastestimator_data', 'horse2zebra')
    else:
        root_dir = os.path.join(os.path.abspath(root_dir), 'horse2zebra')
    os.makedirs(root_dir, exist_ok=True)

    data_compressed_path = os.path.join(root_dir, 'horse2zebra.zip')
    data_folder_path = os.path.join(root_dir, 'images')

    if not os.path.exists(data_folder_path):
        # download
        if not os.path.exists(data_compressed_path):
            print("Downloading data to {}".format(root_dir))
            wget.download('https://people.eecs.berkeley.edu/~taesung_park/CycleGAN/datasets/horse2zebra.zip',
                          root_dir,
                          bar=bar_custom)

        # extract
        print("\nExtracting files ...")
        with zipfile.ZipFile(data_compressed_path, 'r') as zip_file:
            zip_file.extractall(root_dir)
        os.rename(os.path.join(root_dir, 'horse2zebra'), data_folder_path)

    test_a = DirDataset(root_dir=os.path.join(data_folder_path, 'testA'),
                        data_key="A",
                        file_extension='.jpg',
                        recursive_search=False)
    test_b = DirDataset(root_dir=os.path.join(data_folder_path, 'testB'),
                        data_key="B",
                        file_extension='.jpg',
                        recursive_search=False)
    train_a = DirDataset(root_dir=os.path.join(data_folder_path, 'trainA'),
                         data_key="A",
                         file_extension='.jpg',
                         recursive_search=False)
    train_b = DirDataset(root_dir=os.path.join(data_folder_path, 'trainB'),
                         data_key="B",
                         file_extension='.jpg',
                         recursive_search=False)
    outputs = (BatchDataset(datasets=[train_a, train_b], num_samples=[batch_size, batch_size]),
               BatchDataset(datasets=[test_a, test_b], num_samples=[batch_size, batch_size]))
    return outputs