Skip to main content
. Author manuscript; available in PMC: 2022 Apr 30.
Published in final edited form as: Proc Mach Learn Res. 2021 Dec;158:196–208.

Table 7:

Training details for RadBERT-CL Pretraining and Finetuning Stages.

Hyperparameter Pretraining Finetuning
batch-size 128 32
learning-rate 0.1 2e-5
optimizer SGD Adam
temperature (CL) 0.4 -
n_epochs 100 10
beta - [0.9, 0.99]
Aug. Probability 0.2 -