From: Automatic de-identification of French electronic health records: a cost-effective approach exploiting distant supervision and deep learning models
Hyperparameter
Value
Hidden size
256
Batch size
512
Learning rate
0.1
Max_epochs
5
Locked_dropout
0.5
Word_dropout
0.05
Patience
3
Anneal_factor
RNN layers
1
Optimizer
Stochastic gradient descent (SGD)