How deeply does Deep Learning with PyTorch, Second Edition cover rotation?

Implements random 0–2π X-Y rotation with a 4×4 affine matrix, then resamples through PyTorch grid functions.

Rotation

As explained in Deep Learning with PyTorch, Second Edition

A data augmentation transform that turns an image or volume around an axis, here constrained to the X-Y plane for CT data.

Rotation augments CT training data by turning an image or volume around the head-foot axis. The transform changes orientation while keeping the sample representative, helping create useful variation without inventing unrealistic anatomy. section 14.6.1

Listing 14.16 Rotating an image

if 'rotate' in augmentation_dict:
  angle_rad = random.random() * math.pi * 2
  s = math.sin(angle_rad)
  c = math.cos(angle_rad)

  rotation_t = torch.tensor([
    [c, -s, 0, 0],
    [s, c, 0, 0],
    [0, 0, 1, 0],
    [0, 0, 0, 1],
  ])

  transform_t @= rotation_t

The rotation branch samples an angle from 0 to 2π, builds an X-Y rotation matrix with sine and cosine, and multiplies it into the affine transform before resampling. section 14.6.1

How Deep Learning with PyTorch, Second Edition covers it

Cover of Deep Learning with PyTorch, Second Edition

liveBook manning.com

Read Deep Learning with PyTorch, Second Edition on liveBook Get the book at manning.com

Also explained in

liveBook manning.com

Geometry for ProgrammersA rotation is a geometric transformation that turns points around a fixed center, axis, or origin while preserving distances and angles.
liveBook manning.com

CSS in Depth, Second EditionA rotation is a geometric transformation that turns an object around a fixed point or axis.
liveBook manning.com

Math for Web DesignA rotate transform is a CSS transformation operation that rotates an element around its transform origin.

Compare how 4 books explain rotation →

Related concepts

sine functionuses · The sine function is a periodic trigonometric function that maps a real number or angle to a value between −1 and 1.
cosine functionuses · The cosine function is a periodic trigonometric function that maps a real number or angle to a value between −1 and 1.
compositionused by · Composition is a technique for combining simpler functions, components, or operations into a larger structure or behavior.
color scaleconfigures · A color scale is a mapping from data values or categories to colors, often arranged as a continuous gradient or discrete set for visual representation.
positional embeddingused by · A positional embedding is a learned or fixed representation that encodes an element’s position or order in a sequence and is incorporated into its content representation.
radial visualizationused by · A radial visualization arranges graphical marks or data encodings around a center using angular and radial coordinates.

This explainer was generated by Manning's knowledge graph from the text of Deep Learning with PyTorch, Second Edition. The section links point to the parts of the book it draws on.