Source code for mmrotate.core.visualization.palette

# Copyright (c) OpenMMLab. All rights reserved.
import mmcv
import numpy as np


def get_palette(palette, num_classes):
    """Get palette from various inputs.

    Args:
        palette (list[tuple] | str | tuple | :obj:`Color`): palette inputs.
        num_classes (int): the number of classes.

    Returns:
        list[tuple[int]]: A list of color tuples.
    """
    assert isinstance(num_classes, int)

    if isinstance(palette, list):
        dataset_palette = palette
    elif isinstance(palette, tuple):
        dataset_palette = [palette] * num_classes
    elif palette == 'random' or palette is None:
        state = np.random.get_state()
        # random color
        np.random.seed(42)
        palette = np.random.randint(0, 256, size=(num_classes, 3))
        np.random.set_state(state)
        dataset_palette = [tuple(c) for c in palette]
    elif palette == 'dota':
        from mmrotate.datasets import DOTADataset
        dataset_palette = DOTADataset.PALETTE
    elif palette == 'sar':
        from mmrotate.datasets import SARDataset
        dataset_palette = SARDataset.PALETTE
    elif palette == 'hrsc':
        from mmrotate.datasets import HRSCDataset
        dataset_palette = HRSCDataset.PALETTE
    elif palette == 'hrsc_classwise':
        from mmrotate.datasets import HRSCDataset
        dataset_palette = HRSCDataset.CLASSWISE_PALETTE
    elif mmcv.is_str(palette):
        dataset_palette = [mmcv.color_val(palette)[::-1]] * num_classes
    else:
        raise TypeError(f'Invalid type for palette: {type(palette)}')

    assert len(dataset_palette) >= num_classes, \
        'The length of palette should not be less than `num_classes`.'
    return dataset_palette