How can I solve "Error using gpuArray"

Error using gpuArray
An unexpected error occurred during CUDA execution. The CUDA error was:
unspecified launch failure

Error in SeriesNetwork/activations (line 346)
                    X = gpuArray(X);

Error in Untitled (line 115)
trainingFeatures = activations(convnet, trainingSet, featureLayer, ...

this link: https://www.mathworks.com/help/vision/examples/image-category-classification-using-deep-learning.html

.. My GPU is CUDA-capable NVIDIA and compute capability higher 3.0... But I could not solve this problem. I'm glad you helped me.

[Joss Knight]

You'll need to provide more detail about your environment to help us reproduce this issue. What is your GPU, OS, CPU etc.

[nancy mehta]

try restarting your pc

[Ricardo de Azevedo]

I am facing the same problem now training an RNN and have tried both the TdrDelay to longer and the TdrLevel to 0.

Error:

Error using gpuArray/gatherAn unexpected error occurred during CUDA execution. The CUDA error was:CUDA_ERROR_LAUNCH_FAILED

The weird thing is the network trains for a while and then crashes, I can't really tell what triggers it.

(Using Matlab 2019b and latest NVIDIA drivers 441.20 for GTX 1080 Ti, Windows 10, i7-8700K)

Answer 1

[Joss Knight]

Do you have a Pascal card? https://www.mathworks.com/support/bugreports/details/1439741

[Feyzanur Saglam]

thank you for the answer..I havent got pascal card. And my card compute capability lower 6.0 but I dont know why this problem originated..

Answer 2

[Yahya Zakaria mohamed]

https://www.mathworks.com/matlabcentral/answers/299970-reset-gpudevice-does-not-work

Answer 3

[Marius Bledea]

https://uk.mathworks.com/matlabcentral/answers/34052-cuda_error_launch_failed-problem#comment_835725