Enum DeviceType

Types of devices that Sentis uses to run inference on a neural network.

Namespace: Unity.InferenceEngine

Assembly: Unity.InferenceEngine.dll

Syntax

[MovedFrom("Unity.Sentis")]
public enum DeviceType

Remarks

Sentis can run inference on GPU or on CPU. The performance depends on the size of the model and on the type of problem to solve. Smaller models may run faster on CPU.

Examples

When creating a Worker, specify which device to use for model inference.

public ModelAsset model;
Worker worker = new Worker(ModelLoader.Load(model), DeviceType.GPU);

Fields

Name	Description
CPU	Use CPU to run model inference.
GPU	Use GPU to run model inference.