我在机器上安装了tensorflow gpu build。我的tensorflow版本是2.1.0，而cuda是10.0。
当我检查计算机中是否有可用的GPU时，它无法识别图形卡。
这是我的nvidia-smi命令的输出。
mymachine@titanx:~$ nvidia-smi
Fri May 15 00:08:51 2020
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 410.78 Driver Version: 410.78 CUDA Version: 10.0 |
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 GeForce GTX TIT... Off | 00000000:02:00.0 On | N/A |
| 22% 52C P8 19W / 250W | 576MiB / 12211MiB | 0% Default |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| 0 1290 G /usr/lib/xorg/Xorg 182MiB |
| 0 1982 G /opt/teamviewer/tv_bin/TeamViewer 19MiB |
| 0 2483 G compiz 157MiB |
| 0 6030 C python 104MiB |
这是我的错误信息。
mymachine@titanx:~$ python
Python 3.7.4 (default, Aug 13 2019, 20:35:49)
[GCC 7.3.0] :: Anaconda, Inc. on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import tensorflow as tf
2020-05-15 00:15:50.850372: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libnvinfer.so.6'; dlerror: libnvinfer.so.6: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: :/usr/local/cuda/extras/CUPTI/lib64
2020-05-15 00:15:50.850454: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libnvinfer_plugin.so.6'; dlerror: libnvinfer_plugin.so.6: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: :/usr/local/cuda/extras/CUPTI/lib64
2020-05-15 00:15:50.850467: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:30] Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.
>>> tf.test.
tf.test.Benchmark( tf.test.benchmark_config( tf.test.gpu_device_name( tf.test.is_built_with_rocm(
tf.test.TestCase( tf.test.compute_gradient( tf.test.is_built_with_cuda( tf.test.is_gpu_available(
tf.test.assert_equal_graph_def( tf.test.create_local_cluster( tf.test.is_built_with_gpu_support( tf.test.main(
>>> tf.test.is_built_with_gpu_support()
True
>>> tf.test.is_gpu_available()
WARNING:tensorflow:From <stdin>:1: is_gpu_available (from tensorflow.python.framework.test_util) is deprecated and will be removed in a future version.
Instructions for updating:
Use `tf.config.list_physical_devices('GPU')` instead.
2020-05-15 00:16:45.492044: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
2020-05-15 00:16:45.526439: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 3491695000 Hz
2020-05-15 00:16:45.527074: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x55e3d21f7370 initialized for platform Host (this does not guarantee that XLA will be used). Devices:
2020-05-15 00:16:45.527125: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): Host, Default Version
2020-05-15 00:16:45.532644: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcuda.so.1
2020-05-15 00:16:45.623845: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:981] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2020-05-15 00:16:45.624424: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x55e3d22c7d80 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:
2020-05-15 00:16:45.624465: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): GeForce GTX TITAN X, Compute Capability 5.2
2020-05-15 00:16:45.624784: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:981] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2020-05-15 00:16:45.625640: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1555] Found device 0 with properties:
pciBusID: 0000:02:00.0 name: GeForce GTX TITAN X computeCapability: 5.2
coreClock: 1.076GHz coreCount: 24 deviceMemorySize: 11.92GiB deviceMemoryBandwidth: 313.37GiB/s
2020-05-15 00:16:45.625832: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libcudart.so.10.1'; dlerror: libcudart.so.10.1: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: :/usr/local/cuda/extras/CUPTI/lib64
2020-05-15 00:16:45.625967: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libcublas.so.10'; dlerror: libcublas.so.10: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: :/usr/local/cuda/extras/CUPTI/lib64
2020-05-15 00:16:45.626095: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libcufft.so.10'; dlerror: libcufft.so.10: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: :/usr/local/cuda/extras/CUPTI/lib64
2020-05-15 00:16:45.626236: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libcurand.so.10'; dlerror: libcurand.so.10: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: :/usr/local/cuda/extras/CUPTI/lib64
2020-05-15 00:16:45.626389: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libcusolver.so.10'; dlerror: libcusolver.so.10: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: :/usr/local/cuda/extras/CUPTI/lib64
2020-05-15 00:16:45.626527: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libcusparse.so.10'; dlerror: libcusparse.so.10: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: :/usr/local/cuda/extras/CUPTI/lib64
2020-05-15 00:16:45.632796: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7
2020-05-15 00:16:45.632837: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1592] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.
Skipping registering GPU devices...
2020-05-15 00:16:45.632869: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1096] Device interconnect StreamExecutor with strength 1 edge matrix:
2020-05-15 00:16:45.632888: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1102] 0
2020-05-15 00:16:45.632903: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] 0: N
False
>>> tf.__version__
'2.1.0'
所以我尝试安装libnvinfer6和其他实际库（如TensorFlow Official Documentation中所述，但我收到未找到包错误。
mymachine@titanx:~$ sudo apt-get install -y --no-install-recommends libnvinfer6=6.0.1-1+cuda10.1 \
> libnvinfer-dev=6.0.1-1+cuda10.1 \
> libnvinfer-plugin6=6.0.1-1+cuda10.1
[sudo] password for dtai:
Reading package lists... Done
Building dependency tree
Reading state information... Done
E: Unable to locate package libnvinfer6
E: Unable to locate package libnvinfer-dev
E: Unable to locate package libnvinfer-plugin6
我可以看到官方文档提到了我安装cuda 10.0的cuda 10.1。 在Cuda 10.0中安装这些更新的正确命令是什么，以及如何解决“无法找到软件包”的问题？