Application of Convolutional Neural Networks to time domain astrophysics. 2D image analysis of OGLE light curves

Open Access

Table 4

Performance metrics for different models.

Training set	Accuracy	Precision	Recall	F1
Train-8 U	0.890	0.890	0.890	0.891
Train-22 DA	0.911	0.911	0.911	0.912
Train-22 BB	0.908	0.908	0.908	0.909
Train-37 DA	0.913	0.913	0.913	0.915
Train-37 BB	0.911	0.911	0.911	0.912
Train-60 DA	0.916	0.916	0.916	0.918
Train-60 BB	0.916	0.916	0.916	0.917

Notes. Trained on various balanced training sets and evaluated on the same balanced test set.

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.