Table 3 Hyperparameters of the HistoDARE model based on the DINOv2 ViT-L/14 architecture.
From: A histopathology aware DINO model with attention based representation enhancement
Hyperparameter | Value |
|---|---|
Input Size | \(224 \times 224\) |
Patch Size | \(14 \times 14\) |
Embedding Dimension | 1024 |
Transformer Depth (Layers) | 24 |
Number of Attention Heads | 16 |
MLP Ratio | 4.0 |
Drop Path Rate | 0.1 |
Total Number of Parameters | \(\sim\)304M |
FLOPs (for \(224\times 224\) input) | \(\sim\)60G |
