Table 4 Analysis using different optimizers.

From: FasNet: a hybrid deep learning model with attention mechanisms and uncertainty estimation for liver tumor segmentation on LiTS17

Optimizer

Loss

Accuracy

Dice coefficient

Jaccard Index

Precision

Recall

Specificity

F1 Score

AdaGrad

0.0435

0.9531

0.8066

0.7787

0.7236

0.6030

0.9689

0.6163

SGD

0.0299

0.9344

0.7913

0.7743

0.7259

0.8146

0.9394

0.8020

Adam

0.0251

0.9954

0.8766

0.8487

0.8499

0.8560

0.9695

0.8363

RMSProp

0.0435

0.9231

0.7766

0.7587

0.7136

0.7530

0.9289

0.6263

AdaDelta

0.3119

0.9540

0.8043

0.7772

0.7559

0.7850

0.9497

0.7844

Back to article page