AttentiveSamplingStrategy

Enums · AiDotNet.AutoML.NAS

Which candidate AttentiveNAS trains from the pool drawn under a FLOPs target (Wang et al. 2021, “AttentiveNAS: Improving Neural Architecture Search via Attentive Sampling”, Sec. 3.2).

Fields

Field Summary
AttentiveSamplingStrategy BestUp Train the candidate with the highest predicted performance (the Pareto-best).
AttentiveSamplingStrategy WorstUp Train the candidate with the lowest predicted performance (the Pareto-worst).
int value__