nvidia.dali.fn.jitter#

nvidia.dali.fn.jitter(
input,
/,
*,
device=None,
name=None,
bytes_per_sample_hint=[0],
fill_value=0.0,
interp_type=DALIInterpType.INTERP_NN,
mask=1,
nDegree=2,
preserve=False,
seed=-1,
)#

Performs a random Jitter augmentation.

The output images are produced by moving each pixel by a random amount, in the x and y dimensions, and bounded by half of the nDegree parameter.

Supported backends

  • ‘gpu’

Parameters:

input (TensorList ('HWC')) – Input to the operator.

Keyword Arguments:

  • bytes_per_sample_hint (int or list of int, optional, default = [0]) –

    Output size hint, in bytes per sample.

    If specified, the operator’s outputs residing in GPU or page-locked host memory will be preallocated to accommodate a batch of samples of this size.

  • fill_value (float, optional, default = 0.0) – Color value that is used for padding.

  • interp_type (nvidia.dali.types.DALIInterpType, optional, default = DALIInterpType.INTERP_NN) – Type of interpolation used.

  • mask (int or TensorList of int, optional, default = 1) –

    Determines whether to apply this augmentation to the input image.

    Here are the values:

    • 0: Do not apply this transformation.

    • 1: Apply this transformation.

  • nDegree (int, optional, default = 2) – Each pixel is moved by a random amount in the [-nDegree/2, nDegree/2] range

  • preserve (bool, optional, default = False) – Prevents the operator from being removed from the graph even if its outputs are not used.

  • seed (int, optional, default = -1) – Random seed; if not set, one will be assigned automatically.

On this page