Tensor

Also: tensors, n-dimensional array

In deep learning, a tensor is an array with any number of axes (its rank) and a shape listing each axis's size. Frameworks store inputs, weights and activations as tensors and run them on GPUs.

Example

A batch of 32 RGB images at 224×224 is a tensor of shape (32, 3, 224, 224). Stable Diffusion compresses a 512×512×3 image into a 64×64×4 latent tensor of 16,384 values.

See it in the lesson: Stable Diffusion →

Related terms

More in Linear Algebra & Calculus