Okuzeka
Artificial Intelligence & Machine Learning vocabulary · Model Training

Checkpoint

Meaning in artificial intelligence & machine learning, with examples

/ˈtʃekpɔɪnt/nounCommon term

In artificial intelligence & machine learning, checkpoint means: A saved copy of a model's weights, and often its optimizer state, at a certain point in training, so that training can be resumed or that version can be used later.

In Turkish: kontrol noktası, checkpoint, ara kayıt

How it is used

Engineers save, load, resume from and evaluate checkpoints, often every few thousand steps. 'The best checkpoint' is the one with the best validation score. Published models are often called checkpoints too.

Example sentences

  1. The job crashed overnight, but we resumed from the last checkpoint.

  2. Which checkpoint should we ship, the one from step 40k or step 60k?

Common collocations

  • save a checkpoint
  • resume from a checkpoint
  • best checkpoint
  • load the checkpoint
  • weightsThe learned numbers on the connections of a neural network, which decide how strongly each input affects the next layer. Together they hold what the model has learned.
  • trainTo teach a model by showing it many examples and repeatedly adjusting its parameters so that its outputs get closer to the correct answers.
  • epochOne complete pass of the training process through the whole training dataset. Models are usually trained for several epochs.
  • base modelA model that has only been pretrained, before any fine-tuning or instruction training. It continues text well but does not reliably follow instructions or chat.

This word in other fields

The same English word can have a different meaning in another field:

More terms: Model Training

Practise with graded reading

Reading at the right level is the fastest way to make new words stick. Pick free English books by level: