mindspore.dataset.vision.TrivialAugmentWide
- class mindspore.dataset.vision.TrivialAugmentWide(num_magnitude_bins=31, interpolation=Inter.NEAREST, fill_value=0)[source]
Apply TrivialAugmentWide data augmentation method on the input image.
Refer to TrivialAugmentWide: Tuning-free Yet State-of-the-Art Data Augmentation .
Only support 3-channel RGB image.
- Parameters
num_magnitude_bins (int, optional) – The number of different magnitude values, must be greater than or equal to 2. Default:
31
.interpolation (Inter, optional) – Image interpolation method defined by
Inter
. Default:Inter.NEAREST
.fill_value (Union[int, tuple[int, int, int]], optional) – Pixel fill value for the area outside the transformed image, must be in range of [0, 255]. Default:
0
. If int is provided, pad all RGB channels with this value. If tuple[int, int, int] is provided, pad R, G, B channels respectively.
- Raises
TypeError – If num_magnitude_bins is not of type int.
ValueError – If num_magnitude_bins is less than 2.
TypeError – If fill_value is not of type int or tuple[int, int, int].
ValueError – If fill_value is not in range of [0, 255].
RuntimeError – If shape of the input image is not <H, W, C>.
- Supported Platforms:
CPU
Examples
>>> import numpy as np >>> import mindspore.dataset as ds >>> import mindspore.dataset.vision as vision >>> from mindspore.dataset.vision import Inter >>> >>> # Use the transform in dataset pipeline mode >>> data = np.random.randint(0, 255, size=(1, 100, 100, 3)).astype(np.uint8) >>> numpy_slices_dataset = ds.NumpySlicesDataset(data, ["image"]) >>> transforms_list = [vision.TrivialAugmentWide(num_magnitude_bins=31, ... interpolation=Inter.NEAREST, ... fill_value=0)] >>> numpy_slices_dataset = numpy_slices_dataset.map(operations=transforms_list, input_columns=["image"]) >>> for item in numpy_slices_dataset.create_dict_iterator(num_epochs=1, output_numpy=True): ... print(item["image"].shape, item["image"].dtype) ... break (100, 100, 3) uint8 >>> >>> # Use the transform in eager mode >>> data = np.random.randint(0, 255, size=(100, 100, 3)).astype(np.uint8) >>> output = vision.TrivialAugmentWide()(data) >>> print(output.shape, output.dtype) (100, 100, 3) uint8
- Tutorial Examples: