Table 8 Ablation study on different learning rates on CUB_200_2011.

From: Significant feature suppression and cross-feature fusion networks for fine-grained visual classification

Old Layer

New Layer

Accuracy rate (%)

0.0001

0.001

88.7

0.0002

0.002

89.6

0.00001

0.0001

87.2

0.00002

0.0002

87.7

  1. Significant values are in bold.
Back to article page