Skip to main content

Table 7 Training setting for the experiment 2

From: CLeaR: An adaptive continual learning framework for regression tasks

Hyperparameter

Reference figure

Optimizer

Adam

Epochs a in phase 1

256

Epochs p in phase 1

256

Batch size in phase 1

32

Epochs a in phase 2

128

Epochs p in phase 2

256

Batch size in phase 2

16

  1. The parameter epoch a denotes the number of iterations for training the autoencoder and epoch p for training the predictor. The batch size is the size of a mini-batch. Phase 1 and phase 2 indicate the warm-up phase and the update phase respectively