Table 3 The model hyper-parameters settings
From: BPI-MVQA: a bi-branch model for medical visual question answering
Hyper-parameters | batch-size | iteration | Epoch | Learning rate | Dictionary size |
|---|---|---|---|---|---|
ImageCLEF2018 VQA-Med | 64 | 3213 | 38 | 0.0001 | 3000 |
ImageCLEF2019 VQA-Med | 64 | 2698 | 54 | 0.0001 | 3000 |
VQA-RAD | 128 | 885 | 37 | 0.0001 | 3000 |