EDP Sciences logo
Subscriber Authentication Point
Search
Menu
Home All issues Volume 689 (September 2024) A&A, 689 (2024) A289 Full HTML
Open Access

Table 3

Comparison of computational time performances.

Inference step Average time [s]
CPU
Feature computation 1.88 ⋅ 10−1
RF (MD + Features) 2.15 ⋅ 10−4
ATAT (LC + MD + MTA) 6.44 ⋅ 10−3
ATAT (LC + MD + Features + MTA) 1.34 ⋅ 10−2
GPU (2,000 light-curves per batch)
ATAT (LC + MD + MTA) 4.75 ⋅ 10−4
ATAT (LC + MD + Features + MTA) 8.29 ⋅ 10−4
GPU (1 light curve per batch)
ATAT (LC + MD + MTA) 8.14 ⋅ 10−3
ATAT (LC + MD + Features + MTA) 8.45 ⋅ 10−3

Notes. Average computational time per light curve in seconds required to perform the inference step for selected classification models. We used a single core of an AMD EPYC 7662 processor and a NVIDIA A100 GPU for these experiments.

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.