tensorflow 在 gpu 上训练时报如下错误
Failed to get convolution algorithm. This is probably because cuDNN failed to initialize
解决办法
加上如下代码
import os import tensorflow as tf from tensorflow import ConfigProto from tensorflow import InteractiveSession # 注意如果还报错,加在代码开头试试 os.environ['CUDA_VISIBLE_DEVICES'] = '0' # os.environ['CUDA_VISIBLE_DEVICES'] = '/gpu:0' config = ConfigProto() config.gpu_options.allow_growth = True session = InteractiveSession(config=config) # session = tf.Session(config=config)
这是 GPU 显存不够了,上面代码意思是对 GPU 进行按需分配
参考资料:
https://blog.csdn.net/tsyccnh/article/details/102938368
https://www.cnblogs.com/xiaxuexiaoab/p/12388388.html
https://www.jianshu.com/p/e13aa15f35da tensorflow2.0卷积报错:Failed to get convolution algorithm. This is probably because cuDNN failed to i...