MaxTrainingRounds

MaxTrainingRounds

is an option for NetTrain and related functions that specifies the maximum number of rounds of training to do.

Details

  • A training round or "epoch" visits each training example once.
  • With the setting of MaxTrainingRounds->n, training will occur for n rounds.
  • With the default setting of MaxTrainingRounds->Automatic, the number of training rounds to use is calculated to correspond to about 20 seconds of training and is based on the time taken to evaluate one batch of examples. At least one round of training will always be performed.
  • If a setting for TimeGoal is not also specified, MaxTrainingRounds->Automatic will never train for more than 10,000 rounds.

Examples

open allclose all

Basic Examples  (1)

Train a network such that it visits each example exactly once:

Properties & Relations  (1)

When both MaxTrainingRounds and TimeGoal are specified, the shorter of the two will be used (note that this example should be run twice to avoid the initial preprocessing overhead):

Introduced in 2016
 (11.0)