nn.functional.rotate¶

lucid.nn.functional.rotate(input_: Tensor, angle: float, center: tuple[int | float | complex, int | float | complex] | None = None) → Tensor¶

The rotate function rotates an input tensor (typically an image) by a specified angle around a specified center point. This is a common operation in image processing and data augmentation for training neural networks.

Function Signature¶

def rotate(x: Tensor, angle: float, center: tuple[float, float] | None = None) -> Tensor

Parameters¶

x (Tensor): The input image tensor of shape (N, C, H, W), where N is the batch size, C is the number of channels, H is the height, and W is the width.
angle (float): The rotation angle in degrees. Positive values rotate counterclockwise, and negative values rotate clockwise.
center (tuple[float, float] | None, optional): The center of rotation (x, y) in the image. If None, the center of the image ((W/2, H/2)) is used as the rotation center. Default is None.

Returns¶

Tensor: The rotated image tensor with the same shape (N, C, H, W) as the input tensor x.

Usage¶

The rotate function is commonly used in image preprocessing and data augmentation to improve the robustness of neural network models. It allows images to be randomly or deterministically rotated during training.

Example Usage¶

Rotating an image by 45 degrees (center default to image center)

import lucid.nn.functional as F

x = lucid.tensor([[[[1.0, 2.0], [3.0, 4.0]]]])  # Shape: (1, 1, 2, 2)
rotated_image = F.rotate(x, angle=45)
print(rotated_image)

Rotating an image by -30 degrees around a custom center point (2, 2)

import lucid.nn.functional as F

x = lucid.tensor([[[[1.0, 2.0], [3.0, 4.0]]]])  # Shape: (1, 1, 2, 2)
rotated_image = F.rotate(x, angle=-30, center=(2, 2))
print(rotated_image)

Note

Shape Preservation: The shape of the output tensor is identical to the input tensor. The content of the image is rotated within its original size, which may result in some content being clipped.
Center of Rotation: By default, the rotation is centered at (W/2, H/2). If a custom center is provided, the rotation occurs around that point.
Batch Support: The rotate function operates on batches of images, so it can process multiple images simultaneously if N > 1.
Data Augmentation: This function is commonly used for data augmentation to increase the diversity of training datasets and improve model generalization.