Model performance estimation method and device and storage medium

The invention provides a model performance estimation method and device and a storage medium, and belongs to the technical field of computer equipment, and the method comprises the steps: obtaining sample data which comprises graph structure data of a plurality of sample neural network models and op...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
1. Verfasser: HE YUANDUO
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The invention provides a model performance estimation method and device and a storage medium, and belongs to the technical field of computer equipment, and the method comprises the steps: obtaining sample data which comprises graph structure data of a plurality of sample neural network models and operation time consumption corresponding to the sample neural network models; according to the multiple pieces of graph structure data and the corresponding operation time consumption, training a preset graph convolutional neural network model until the preset graph convolutional neural network model converges to obtain a performance estimation model; obtaining a to-be-detected neural network model, and performing graph structure processing on the neural network model to generate to-be-detected graph structure data; and inputting the to-be-tested graph structure data into the performance estimation model to obtain a performance estimation result of the to-be-detected neural network model. According to the method, the