notes

Personal notes
git clone git://git.laack.co/notes.git
Log | Files | Refs

EarlyStopping.md (472B)


      1 # Early Stopping
      2 
      3 ML D3
      4 
      5 **Definition:** Early stopping is the process of stopping a model early in training (assuming it uses GD or something akin to that) as a form of regularization.
      6 
      7 Early stopping decreases overfitting by stopping once a certain prediction error threshold is met. This also reduces time to train.
      8 
      9 Using sklearn, we can use partial_fit along with an epoch (pass) counter and loss calculation to determine if we are close enough to some goal to stop.