Graph View

loss

1 min read

Loss functions, equations and use in pytorch explained (+ creating custom loss functions).

Don’t seperate loss functions from models. Blogpost

Loss functions typically measure how well a model is able to recreate a pattern - they are syntactic criterions.


zero-one loss
quadratic loss
negative log-likelihood loss
cross-entropy loss
mean absolute error
mean squared error
focal loss
dice loss


Backlinks

  • KL-divergence
  • batch gradient descent
  • cross-entropy loss
  • energy-based models
  • generalization error
  • meta learning
  • negative log-likelihood loss
  • quadratic loss
  • supervised learning
  • surprise

Created with Quartz v4.5.2 © 2026