[python] UnknownError: 2 root error(s) found. (0) Unknown: Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try looking to see if a warning log message was printed above.

Error

UnknownError: 2 root error(s) found. (0) Unknown: Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try looking to see if a warning log message was printed above. [[node sequential/conv1d/conv1d (defined at <ipython-input-170-155e8bf0429b>:2) ]] [[sparse_categorical_crossentropy/Reshape_1/_12]] (1) Unknown: Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try looking to see if a warning log message was printed above. [[node sequential/conv1d/conv1d (defined at <ipython-input-170-155e8bf0429b>:2) ]] 0 successful operations. 0 derived errors ignored. [Op:__inference_train_function_4698] Function call stack: train_function -> train_function

 

Solution

Because of GPU memory ( too small )

① Change the prameter
② Restart the program
③ Kill the other program that use the GPU
④ If you use the muti-GPU, select the GPU to use
→ https://www.tensorflow.org/guide/gpu?hl=ko