tfl.conditional_pwl_calibration.default_keypoint_input_parameters

Helper creating default keypoint_input_parameters.

tfl.conditional_pwl_calibration.default_keypoint_input_parameters(
    num_keypoints: Optional[int] = None,
    keypoints: Optional[Sequence[float]] = None,
    units: int = 1
) -> Optional[tf.Tensor]

Primarily used for testing.

num_keypoints number of keypoints. If provided, keypoints will be equally spaced.
keypoints sequence of increasing keypoints.
units number of parallel calibrations on one input.

A tensor with a shape of (1, units, input_param_size) or (1, units, input_param_size).

ValueError if parsing failed.