Problem with CuDNN initialization

Dear Technical help,

I have installed lambda-stack in my system (ubuntu) and installed Pycharm with the “inherit global site-packages” setting to use the lambda-stack environment.

However, when I try to run some demo Tensorflow jobs, I got the error of CuDNN initialization (figure attached). Do you have any suggestions for fixing this problem?

Many thanks!

Try adding the following after importing Tensorflow:

for device in tf.config.experimental.list_physical_devices('GPU'):
    tf.config.experimental.set_memory_growth(device, True)
1 Like