Thread group count is above the maximum allowed limit

Hi, I am trying out a pre-trained model (https://github.com/onnx/models/tree/master/vision/object_detection_segmentation/duc)

But I keep getting an error that the thread group is above the maximum allowed limit.
Any tips or direction on how I could fix it?

Error

Code (CSharp):

1. Thread group count is above the maximum allowed limit. Maximum allowed thread group count is 65535.
2. UnityEngine.ComputeShader:Dispatch(Int32, Int32, Int32, Int32)
3. Barracuda.ComputeFunc:Dispatch(Int32, Int32, Int32) (at Library/PackageCache/com.unity.barracuda@0.4.0-preview/Barracuda/Core/Backends/BarracudaReferenceCompute.cs:1422)
4. Barracuda.ComputeFunc:Dispatch(Int32[]) (at Library/PackageCache/com.unity.barracuda@0.4.0-preview/Barracuda/Core/Backends/BarracudaReferenceCompute.cs:1408)
5. Barracuda.ComputeKernel:Dispatch() (at Library/PackageCache/com.unity.barracuda@0.4.0-preview/Barracuda/Core/Backends/BarracudaCompute.cs:435)
6. Barracuda.ComputeOps:Softmax(Tensor) (at Library/PackageCache/com.unity.barracuda@0.4.0-preview/Barracuda/Core/Backends/BarracudaCompute.cs:847)
7. Barracuda.StatsOps:Barracuda.IOps.Softmax(Tensor) (at Library/PackageCache/com.unity.barracuda@0.4.0-preview/Barracuda/Core/Backends/StatsOps.cs:209)
8. Barracuda.<ExecuteAsync>d__27:MoveNext() (at Library/PackageCache/com.unity.barracuda@0.4.0-preview/Barracuda/Core/Backends/GenericWorker.cs:587)
9. Barracuda.GenericWorker:Execute() (at Library/PackageCache/com.unity.barracuda@0.4.0-preview/Barracuda/Core/Backends/GenericWorker.cs:117)
10. Barracuda.GenericWorker:Execute(IDictionary`2) (at Library/PackageCache/com.unity.barracuda@0.4.0-preview/Barracuda/Core/Backends/GenericWorker.cs:105)
11. <Working>d__10:MoveNext() (at Assets/Scripts/ModelTester.cs:59)
12. UnityEngine.SetupCoroutine:InvokeMoveNext(IEnumerator, IntPtr)
13.  

Script

Code (CSharp):

1. using Barracuda;
2. using System;
3. using System.Collections;
4. using System.Collections.Generic;
5. using UnityEngine;
6.  
7. public class ModelTester : MonoBehaviour
8. {
9.     private const string INPUT_NAME = "data";
10.  
11.     [SerializeField]
12.     private NNModel modelFile;
13.  
14.     [SerializeField]
15.     private Texture2D texture;
16.  
17.     private IWorker worker;
18.  
19.     private void Start() {
20.         Model model = ModelLoader.Load(this.modelFile, false);
21.  
22.         this.worker = WorkerFactory.CreateWorker(WorkerFactory.Type.ComputePrecompiled, model);
23.         StartCoroutine(Working());
24.     }
25.  
26.     private IEnumerator Working() {
27.         Debug.Log("Starting");
28.         yield return new WaitForSeconds(1);
29.  
30.         using (var tensor = new Tensor(texture)) {
31.             Dictionary<string, Tensor> inputs = new Dictionary<string, Tensor>();
32.  
33.             try {
34.                 inputs.Add(INPUT_NAME, tensor);
35.  
36.                 this.worker.Execute(inputs);
37.             } catch (Exception ex) {
38.                 Debug.LogError(ex);
39.             }
40.  
41.             // Do stuff with the result
42.             Debug.Log("Done");
43.         }
44.     }
45. }

 

Thanks! will do that