python - 如何启用 GPU 在 Anaconda 环境中工作？我设置了一切

Question

我使用带有 Nvidia Tesla T4 的 Ubuntu 18 VM。我安装了支持 GPU 的更正 tensorflow 版本 1.15，但仍然无法让 GPU 在 jupyter notebook 中工作。如何处理这个问题..我相信驱动设置成功

+-----------------------------------------------------------------------------+
| NVIDIA-SMI 440.33.01    Driver Version: 440.33.01    CUDA Version: 10.2     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  Tesla P4            On   | 00000000:00:04.0 Off |                    0 |
| N/A   35C    P8     7W /  75W |      0MiB /  7611MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID   Type   Process name                             Usage      |
|=============================================================================|
|  No running processes found                                                 |
+-----------------------------------------------------------------------------+
(base) koruplato@instance-1:~$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2019 NVIDIA Corporation
Built on Wed_Oct_23_19:24:38_PDT_2019
Cuda compilation tools, release 10.2, V10.2.89

据说

print("Num GPUs Available: ", len(tf.config.experimental.list_physical_devices('GPU')))

Num GPUs Available:  0```

标签： pythontensorflowgpu

Answer 1

据我所知，您拥有 CUDA 10.2。TensorFlow 仅适用于 CUDA 10.0，如下所述：https ://www.tensorflow.org/install/gpu 。

您还需要获取 cuDNN SDK (>=7.4.1)

希望有帮助