Question

我运行构建tensorflow / tensorflow，这是来自Docker Hub的Docker映像的CPU版本。它建立在没有NVIDIA GPU的计算机上。在python中使用tensorflow库时会引发一些错误。

2020-06-13 07:47:05.279642: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libcuda.so.1'; dlerror: libcuda.so.1: cannot open shared object file: No such file or directory
2020-06-13 07:47:05.282730: E tensorflow/stream_executor/cuda/cuda_driver.cc:313] failed call to cuInit: UNKNOWN ERROR (303)
2020-06-13 07:47:05.287002: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:156] kernel driver does not appear to be running on this host (f7be0d154c64): /proc/driver/nvidia/version does not exist
2020-06-13 07:47:05.297515: I tensorflow/core/platform/cpu_feature_guard.cc:143] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
2020-06-13 07:47:05.566521: I tensorflow/core/platform/profile_utils/cpu_utils.cc:102] CPU Frequency: 2399995000 Hz
2020-06-13 07:47:05.571228: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x7f3080000b20 initialized for platform Host (this does not guarantee that XLA will be used). Devices:
2020-06-13 07:47:05.571293: I tensorflow/compiler/xla/service/service.cc:176]   StreamExecutor device (0): Host, Default Version
2020-06-13 07:47:10.799206: W tensorflow/core/framework/cpu_allocator_impl.cc:81] Allocation of 614400000 exceeds 10% of free system memory.
Killed

我想知道是否也需要NVIDIA GPU来用于带有CPU版本的tensorflow / tensorflow Docker映像？请帮助我使用tensorflow CPU构建Docker映像。

Answer 1

这些步骤运行正常：

docker run -it -d --name tensor tensorflow/tensorflow
移动到新创建的容器docker exec -it tensor bash
python
>> import tensorflow

张量流/张量流Docker映像是否需要NVIDIA GPU？

1 个答案: