Table 2 Summary of classification...

Table 2

Summary of classification performance for four CNN architectures trained to predict stress-induced flowering success, comparing the Adam and Adamax optimizers, reported over 10-fold cross validation for accuracy, precision, recall and F1-score

Architecture	Optimizer	Accuracy (mean ± std)	Precision (mean ± std)	Recall (mean ± std)	F1-score (mean ± std)
VGG-16	Adam	0.38 ± 0.04	0.13 ± 0.03	0.33 ± 0.05	0.18 ± 0.04
VGG-16	Adamax	0.78 ± 0.02	0.77 ± 0.02	0.77 ± 0.02	0.77 ± 0.02
ResNet-50V2	Adam	0.86 ± 0.01	0.87 ± 0.01	0.86 ± 0.01	0.86 ± 0.01
ResNet-50V2	Adamax	0.83 ± 0.02	0.88 ± 0.02	0.81 ± 0.02	0.83 ± 0.02
MobileNetV2	Adam	0.38 ± 0.05	0.13 ± 0.03	0.33 ± 0.05	0.18 ± 0.04
MobileNetV2	Adamax	0.81 ± 0.02	0.86 ± 0.02	0.80 ± 0.02	0.81 ± 0.02
InceptionV3^a	Adam	0.75 ± 0.02	0.75 ± 0.02	0.74 ± 0.02	0.74 ± 0.02
InceptionV3^a	Adamax	0.90 ± 0.01	0.92 ± 0.01	0.88 ± 0.01	0.89 ± 0.01

Architecture	Optimizer	Accuracy (mean ± std)	Precision (mean ± std)	Recall (mean ± std)	F1-score (mean ± std)
VGG-16	Adam	0.38 ± 0.04	0.13 ± 0.03	0.33 ± 0.05	0.18 ± 0.04
VGG-16	Adamax	0.78 ± 0.02	0.77 ± 0.02	0.77 ± 0.02	0.77 ± 0.02
ResNet-50V2	Adam	0.86 ± 0.01	0.87 ± 0.01	0.86 ± 0.01	0.86 ± 0.01
ResNet-50V2	Adamax	0.83 ± 0.02	0.88 ± 0.02	0.81 ± 0.02	0.83 ± 0.02
MobileNetV2	Adam	0.38 ± 0.05	0.13 ± 0.03	0.33 ± 0.05	0.18 ± 0.04
MobileNetV2	Adamax	0.81 ± 0.02	0.86 ± 0.02	0.80 ± 0.02	0.81 ± 0.02
InceptionV3^a	Adam	0.75 ± 0.02	0.75 ± 0.02	0.74 ± 0.02	0.74 ± 0.02
InceptionV3^a	Adamax	0.90 ± 0.01	0.92 ± 0.01	0.88 ± 0.01	0.89 ± 0.01

Note(s)

Selected as the best model

[ViewLarge]

Sharing Unavailable