tensorflow.python.framework.errors_impl.InternalError: Blas GEMM launch failed 해결 방법

Tnesorflow 2.0 and 2.1

import tensorflow as tf
tf.config.gpu.set_per_process_memory_growth(True)

 

Tensorflow 2.2+

import tensorflow as tf
gpus = tf.config.experimental.list_physical_devices('GPU')
for gpu in gpus:
  tf.config.experimental.set_memory_growth(gpu, True)

이 에러가 뜰 때 맨 밑에

Function call stack:
train_function

가 떴었는데 이건 의미 없었다.

 

 

출처:

https://stackoverflow.com/questions/34199233/how-to-prevent-tensorflow-from-allocating-the-totality-of-a-gpu-memory/34200194

How to prevent tensorflow from allocating the totality of a GPU memory?