Euclid preparation - XXXIII. Characterization of convolutional neural networks for the identification of galaxy-galaxy strong-lensing events

Home

All issues

Volume 681 (January 2024)

A&A, 681 (2024) A68

Full HTML

Open Access

Table B.1

Summary of the performance of the VGG-like network, the IncNet, and the ResNet in classifying the objects of the four selections of images in the I_E band.

VGG-like network

	S1		S2		S3		S4
Class	0	1	0	1	0	1	0	1

Precision	0.95	0.98	0.94	0.97	0.92	0.94	0.79	0.89

Recall	0.98	0.94	0.98	0.94	0.94	0.92	0.90	0.77

F1-score	0.96	0.96	0.96	0.96	0.93	0.93	0.84	0.83

Accuracy	0.96		0.96		0.93		0.84

AUC	0.77		0.58		0.88		0.57

Inception Network

	S1		S2		S3		S4
Class	0	1	0	1	0	1	0	1

Precision	0.97	1.0	0.97	0.96	0.94	0.93	0.84	0.90

Recall	1.0	0.96	0.96	0.97	0.93	0.94	0.91	0.83

F1-score	0.98	0.98	0.96	0.96	0.93	0.94	0.87	0.86

Accuracy	0.98		0.96		0.94		0.87

AUC	0.92		0.88		0.90		0.81

Residual Network

	S1		S2		S3		S4
Class	0	1	0	1	0	1	0	1

Precision	0.93	0.97	0.90	0.92	0.86	0.89	0.71	0.84

Recall	0.97	0.92	0.92	0.89	0.89	0.85	0.87	0.66

F1-score	0.95	0.94	0.91	0.91	0.88	0.87	0.78	0.74

Accuracy	0.95		0.91		0.87		0.76

AUC	0.81		0.85		0.79		0.70

Notes. The precision, recall, and F1-score are evaluated on the class of the nonlenses (0) and of the lenses (1) separately, while accuracy and AUC are global quantities.

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.