unable to create StreamExecutor for CUDA:0: failed initializing StreamExecutor for CUDA device...

最新推荐文章于 2023-04-04 23:20:09 发布

原创

最新推荐文章于 2023-04-04 23:20:09 发布 · 2.3k 阅读

0 ·

CC 4.0 BY-SA版权

文章标签：

#gpu #cuda #tensorflow #神经网络 #ubuntu

本文档针对在Ubuntu 16.04环境下，使用CUDA 10.0和CUDNN 7.6.5时遇到的TensorFlow-GPU 2.0初始化CUDA:0失败的问题进行讨论。错误发生于网络初始化前，解决方案是在导入包之后添加特定代码段。参考了多个资源以解决此卷积算法失败的报错。

环境

Ubuntu: 16.04
CUDA: 10.0
CUDNN: 7.6.5

执行情况

$ python3 test.py
2020-10-13 15:20:14.602211: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcuda.so.1
2020-10-13 15:20:14.605874: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:1006] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2020-10-13 15:20:14.606291: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1618] Found device 0 with properties: 
name: Tesla V100-SXM2-32GB major: 7 minor: 0 memoryClockRate(GHz): 1.53
pciBusID: 0000:00:08.0
2020-10-13 15:20:14.606480: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.0
2020-10-13 15:20:14.607907: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10.0
2020-10-13 15:20:14.609115: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcufft.so.10.0
2020-10-13 15:20:14.609426: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcurand.so.10.0
2020-10-13 15:20:14.611090: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusolver.so.10.0
2020-10-13 15:20:14.612343: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusparse.so.10.0
2020-10-13 15:20:14.616370: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7
2020-10-13 15:20:14.616475: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:1006] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2020-10-13 15:20:14.616929: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:1006] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2020-10-13 15:20:14.617299: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1746] Adding visible gpu devices: 0
2020-10-13 15:20:14.617549: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
2020-10-13 15:20:14.625251: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 2499995000 Hz
2020-10-13 15:20:14.625774: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x560b01298670 executing computations on platform Host. Devices: