# Loss curve

> A plot of loss as a function of the number of training iterations. Loss curves can help you determine when your model is converging or overfitting.[^1]

A plot of [loss](https://wiki.g15e.com/pages/Loss%20(machine%20learning.txt)) as a function of the number of training [iterations](https://wiki.g15e.com/pages/Iteration%20(machine%20learning.txt)). Loss curves can help you determine when your model is [converging](https://wiki.g15e.com/pages/Convergence%20(machine%20learning.txt)) or [overfitting](https://wiki.g15e.com/pages/Overfitting.txt).[^1]

Loss curves can plot all of the following types of loss:[^1]

- [Training loss](https://wiki.g15e.com/pages/Training%20loss.txt)
- [Validation loss](https://wiki.g15e.com/pages/Validation%20loss.txt)
- [Test loss](https://wiki.g15e.com/pages/Test%20loss.txt)

## See also

- [Generalization curve](https://wiki.g15e.com/pages/Generalization%20curve.txt)

## Footnotes

[^1]: https://developers.google.com/machine-learning/glossary#loss-curve