a CUDA runtime call was likely performed without using a StreamExecutor context

Question

a CUDA runtime call was likely performed without using a StreamExecutor context

ruotianluo opened this issue 8 years ago · comments

I got error like this, which seems to be caused by conflicts between caffe and tensorflow:

current context was not created by the StreamExecutor cuda_driver API: 0x2cc2820; a CUDA runtime call was likely performed without using a StreamExecutor context

However, I didn't find the way to solve it.

Ruotian(RT) Luo · Answer 1 · Wed Jan 25 2017 00:36:37 GMT+0800 (China Standard Time)

Solve it by replace the gpu tensorflow with cpu tensorflow

Xiaofan Xu · Answer 2 · Wed Mar 22 2017 23:36:53 GMT+0800 (China Standard Time)

@ruotianluo Hi I got the same error, but in the code I am using the cpu mode for tensorflow.
with tf.device('/cpu:0'):