Model training method and device and electronic equipment

The invention discloses a model training method and device and electronic equipment, and the method comprises the steps: obtaining an initial picture containing preset information and a target Gaussian sample, carrying out the forward noise enhancement processing of the initial picture, obtaining a...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: WANG XINGKE, HU JING, CAI XIN
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 discloses a model training method and device and electronic equipment, and the method comprises the steps: obtaining an initial picture containing preset information and a target Gaussian sample, carrying out the forward noise enhancement processing of the initial picture, obtaining a first loss value corresponding to the forward training, carrying out the iterative training of a training model based on the first loss value and the target Gaussian sample, and obtaining a second loss value corresponding to the forward training; and determining a second loss value corresponding to the training model and a preset function, stopping iterative training when the preset function converges, and taking the training model corresponding to the preset function as a target model. According to the method, the initial picture is subjected to forward training, so that the distribution difference between the output picture data and the initial picture can be determined, and then the target Gaussian sample is sub