AI Glossary

Batch size

Also called batch, mini-batch

What does “Batch size” mean?

Batch size: A batch is the group of training examples a model processes together before each weight update, and batch size is the number of examples in that group. Larger batches make better use of hardware, while the chosen size also affects how stable and efficient training is.

Batch size visual guide: A batch is the group of training examples a model processes together before each weight update, and batch size is the number of examples in that group. Larger batches make better use of hardware, while the chosen size also affects how stable and efficient training is.
Visual guide Open full size (new tab)

What is an example of “Batch size”?

If a dataset has 10,000 examples and the batch size is 100, one epoch takes 100 weight updates.

Keep learning

See also

EpochTermGradient descentTermLearning rateTermHyperparameterTerm

Same topic

CheckpointTermBackpropagationTermComputeTermCurriculum learningTermData labelingTerm