只需安装Cuda即可使用我的GPU来获取Tensorflow。我按照https://www.tensorflow.org/install/gpu上的说明进行了操作,并在https://docs.nvidia.com/cuda/cuda-installation-guide-microsoft-windows/上运行了Cuda测试(deviceQuery和bandwidthTest),都很好。
当我打开PyCharm并运行以下代码时
import tensorflow as tf
print(tf.test.is_gpu_available(
cuda_only=False,
min_cuda_compute_capability=None
))
计算结果为True
,但出现以下错误消息:
2019-04-12 16:45:48.713478: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2
2019-04-12 16:45:48.918686: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1433] Found device 0 with properties:
name: GeForce GTX 1070 major: 6 minor: 1 memoryClockRate(GHz): 1.7085
pciBusID: 0000:01:00.0
totalMemory: 8.00GiB freeMemory: 6.64GiB
2019-04-12 16:45:48.919057: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1512] Adding visible gpu devices: 0
True
2019-04-12 16:45:49.497524: I tensorflow/core/common_runtime/gpu/gpu_device.cc:984] Device interconnect StreamExecutor with strength 1 edge matrix:
2019-04-12 16:45:49.497813: I tensorflow/core/common_runtime/gpu/gpu_device.cc:990] 0
2019-04-12 16:45:49.497956: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1003] 0: N
2019-04-12 16:45:49.498209: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] Created TensorFlow device (/device:GPU:0 with 6389 MB memory) -> physical GPU (device: 0, name: GeForce GTX 1070, pci bus id: 0000:01:00.0, compute capability: 6.1)
这些是什么意思,我应该关注什么?
(正在运行Windows 10,Python 3.6.8,Cuda Toolkit 10.0,Tensorflow 1.13.1)