tensors

Matrices with depth

Tensors are mathematical objects that generalize scalars, vectors, and matrices to higher dimensions.

In physics and engineering, tensors describe quantities that transform in specific ways under coordinate changes—stress, strain, and electromagnetic fields are tensor quantities.

In machine learning, tensor usually means a multidimensional array of numbers: The neural network processes tensors of varying shapes.

The word has become common in programming through frameworks like TensorFlow.

mathematics

matrices

Rectangular numerical arrangements

rank · indices · contraction · transformation · dimension · basis · components · covariant · invariant · eigenvalues

computing

arrays

Ordered element collections

neural networks · deep learning · operations · broadcasting · reshape · gradients · backpropagation · batch · inference · parallelization

physics

stress

Pressure, mental or vocal

curvature · metric · field · relativity · spacetime · elasticity · deformation · Einstein · gravitation · momentum

notation

subscripts

Below-baseline text markers

coordinate · contraction · raising · lowering

Categories
Opposites
mathematical complexity/dimensionality

tensorsscalars

Direction-free numerical values

constants · real numbers · numbers · invariants · vectors

Phrases
Family

tensor

Familiarity
Top 1,000,000
Profile
Used worldwide · Timeless · Abstract · Rated G