CUDNN_STATUS_EXECUTION_FAILED

I am enrolled in the Nvidia Course: Rendered Image Denoising using Autoencoders. I tried to recreate the environment found in the course, on my dual boot Ubuntu 16.04 OS. There is an executable in the course named nvdenoise. When I try to run nvdenoise in my setup, it gives me the following message:

Cuda error: CUDNN_STATUS_EXECUTION_FAILED

Does anyone have any suggestions on how to solve this issue?

have you solved this?
same problem here

This is usually caused by a failure to launch some cuDNN kernel on the GPU, which can occur for multiple reasons.
To correct: check that the hardware, an appropriate version of the driver, and the cuDNN library are correctly installed.
Please refer to below link for more details.
https://docs.nvidia.com/deeplearning/sdk/cudnn-api/index.html#cudnnStatus_t

In order to avoid system level dependencies will recommend to use NGC container.

Thanks