blocks|key|1611495|text|Tensorflow+2.1%2B|type|unstyled|depth|inlineStyleRanges|entityRanges|data|1611496|到底怎么回事？|1611497|对于新的Tensorflow+2.1发行版，默认的tensorflow+pip包包含TF的CPU和GPU版本。在以前的TF版本中，找不到CUDA库会发出错误并引发异常，而现在库会动态地搜索正确的CUDA版本，如果找不到，就会发出警告(开头的W表示警告，错误具有E+(或表示致命错误的F+)，然后回到CPU--仅限模式)。实际上，这也是在日志中写成警告后的信息消息(请注意，如果您有更高的最低日志级别，默认情况下，您可能看不到info消息)。完整的原木是(重点是地雷)：|offset|length|style|CODE|BOLD|1611498|2020-01-20+12:27:44.554767:+W+tensorflow/stream_executor/platform/default/dso_loader.cc:55]无法加载动态库cudart64+64_101.dll；dlerror:+cudart64_101.dll未找到
2020-01-20+12:27:44.554964:如果你没有在你的机器上安装GPU，我tensorflow/stream_executor/cuda/cudart_stub.cc:29]+忽略上面的cudart+dlerror。|blockquote|1611499|我该担心吗？我该怎么解决呢？|1611500|如果你在你的机器上没有一个支持CUDA的GPU，或者你不关心没有GPU加速，就不用担心了。另一方面，如果您安装了tensorflow并希望GPU加速，请检查您的CUDA安装(TF+2.1需要CUDA+10.1，而不是10.2或10.0)。|1611501|如果您只想消除警告，您可以使用调整TF的测井水平来抑制警告，但这可能是过分的，因为它将沉默所有的警告。|1611502|Tensorflow+1.X或2.0：|1611503|您的CUDA设置已损坏，请确保已安装正确的版本。|1611504|entityMap|0|LINK|mutability|MUTABLE|url|https://github.com/tensorflow/tensorflow/releases/tag/v2.1.0|1|https://stackoverflow.com/questions/35911252/disable-tensorflow-debugging-information^0|0|0|P|A|3C|1|3M|1|3X|1|2|J|0|0|6R|K|0|0|2S|4|0|F|9|1|0|0|0^^$0|@$1|2|3|4|5|6|7|17|8|@]|9|@]|A|$]]|$1|B|3|C|5|6|7|18|8|@]|9|@]|A|$]]|$1|D|3|E|5|6|7|19|8|@$F|1A|G|1B|H|I]|$F|1C|G|1D|H|J]|$F|1E|G|1F|H|J]|$F|1G|G|1H|H|J]]|9|@$F|1I|G|1J|1|1K]]|A|$]]|$1|K|3|L|5|M|7|1L|8|@$F|1M|G|1N|H|J]]|9|@]|A|$]]|$1|N|3|O|5|6|7|1O|8|@]|9|@]|A|$]]|$1|P|3|Q|5|6|7|1P|8|@$F|1Q|G|1R|H|J]]|9|@]|A|$]]|$1|R|3|S|5|6|7|1S|8|@]|9|@$F|1T|G|1U|1|1V]]|A|$]]|$1|T|3|U|5|6|7|1W|8|@]|9|@]|A|$]]|$1|V|3|W|5|6|7|1X|8|@]|9|@]|A|$]]|$1|X|3|-4|5|6|7|1Y|8|@]|9|@]|A|$]]]|Y|$Z|$5|10|11|12|A|$13|14]]|15|$5|10|11|12|A|$13|16]]]]

<h1>Tensorflow 2.1+</h1>
<h2>What's going on?</h2>
With the <a href="https://github.com/tensorflow/tensorflow/releases/tag/v2.1.0" rel="noreferrer">new Tensorflow 2.1 release</a>, the default <code>tensorflow</code> pip package contains both CPU and GPU versions of TF. In previous TF versions, not finding the CUDA libraries would emit an error and raise an exception, while now the library dynamically searches for the correct CUDA version and, if it doesn't find it, emits the warning (The W in the beginning stands for warnings, errors have an E (or F for fatal errors) and falls back to CPU-only mode. In fact, this is also written in the log as an info message right after the warning (do note that if you have a higher minimum log level that the default, you might not see info messages). The full log is (emphasis mine):
<blockquote>
2020-01-20 12:27:44.554767: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'cudart64_101.dll'; dlerror: cudart64_101.dll not found
2020-01-20 12:27:44.554964: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
</blockquote>
<h2>Should I worry? How do I fix it?</h2>
If you don't have a CUDA-enabled GPU on your machine, or if you don't care about not having GPU acceleration, no need to worry. If, on the other hand, you installed tensorflow and wanted GPU acceleration, check your CUDA installation (TF 2.1 requires CUDA 10.1, not 10.2 or 10.0).
If you just want to get rid of the warning, you can <a href="https://stackoverflow.com/questions/35911252/disable-tensorflow-debugging-information">adapt TF's logging level</a> to suppress warnings, but that might be overkill, as it will silence all warnings.
<hr />
<h1>Tensorflow 1.X or 2.0:</h1>
Your CUDA setup is broken, ensure you have the correct version installed.

blocks|key|767233|text|要在TensorFlow+2.1中安装GPU支持的先决条件：|type|unstyled|depth|inlineStyleRanges|entityRanges|data|767234|安装最新的GPU驱动程序。|ordered-list-item|767235|安装库达10.1。|offset|length|767236|如果CUDA安装程序报告“您正在安装较旧的驱动程序版本”，则可能希望选择自定义安装并取消某些组件。实际上，请注意，与CUDA捆绑的软件(包括GeForce体验、PhysX、显示驱动程序和Visual集成)并不是TensorFlow所需要的。|unordered-list-item|767237|还请注意，除非您是从源代码构建的，否则TensorFlow需要一个特定版本的CUDA工具包；对于TensorFlow+2.1和2.2，这是目前的10.1版本。|767238|767239|​|767240|安装cuDNN。|767241|下载cuDNN+v7.6.4+for+CUDA+10.1.这将要求您注册NVIDIA开发计划。|767242|解压缩到合适的位置并将bin目录添加到您的路径中。|767243|767244|767245|通过pip+install+tensorflow安装tensorflow。|style|CODE|767246|你可能需要重新启动你的电脑。|767247|entityMap|0|LINK|mutability|MUTABLE|url|https://developer.nvidia.com/cuda-10.1-download-archive-base|1|https://developer.nvidia.com/rdp/cudnn-archive|2|https://stackoverflow.com/a/51112550/604687^0|0|0|2|6|0|1|1|1|0|0|1|0|7|1|1|1|0|0|2|M|0|1|C|2|0^^$0|@$1|2|3|4|5|6|7|1G|8|@]|9|@]|A|$]]|$1|B|3|C|5|D|7|1H|8|@]|9|@]|A|$]]|$1|E|3|F|5|D|7|1I|8|@]|9|@$G|1J|H|1K|1|1L]]|A|$]]|$1|I|3|J|5|K|7|1M|8|@]|9|@]|A|$]]|$1|L|3|M|5|K|7|1N|8|@]|9|@]|A|$]]|$1|N|3|-4|5|6|7|1O|8|@]|9|@]|A|$]]|$1|O|3|P|5|6|7|1P|8|@]|9|@]|A|$]]|$1|Q|3|R|5|D|7|1Q|8|@]|9|@]|A|$]]|$1|S|3|T|5|D|7|1R|8|@]|9|@$G|1S|H|1T|1|1U]]|A|$]]|$1|U|3|V|5|D|7|1V|8|@]|9|@]|A|$]]|$1|W|3|-4|5|6|7|1W|8|@]|9|@]|A|$]]|$1|X|3|P|5|6|7|1X|8|@]|9|@]|A|$]]|$1|Y|3|Z|5|D|7|1Y|8|@$G|1Z|H|20|10|11]]|9|@]|A|$]]|$1|12|3|13|5|D|7|21|8|@]|9|@$G|22|H|23|1|24]]|A|$]]|$1|14|3|-4|5|6|7|25|8|@]|9|@]|A|$]]]|15|$16|$5|17|18|19|A|$1A|1B]]|1C|$5|17|18|19|A|$1A|1D]]|1E|$5|17|18|19|A|$1A|1F]]]]

To install the prerequisites for GPU support in TensorFlow 2.1:

<ol>
<li>Install your latest GPU drivers.</li>
<li>Install <a href="https://developer.nvidia.com/cuda-10.1-download-archive-base" rel="noreferrer">CUDA 10.1</a>.

<ul>
<li>If the CUDA installer reports "you are installing an older driver version", you may wish to choose a custom installation and deselect some components. Indeed, note that software bundled with CUDA including GeForce Experience, PhysX, a Display Driver, and Visual Studio integration are not required by TensorFlow.</li>
<li>Also note that TensorFlow requires a specific version of the CUDA Toolkit unless you build from source; for TensorFlow 2.1 and 2.2, this is currently version 10.1.</li>
</ul></li>
<li>Install cuDNN.

<ol>
<li><a href="https://developer.nvidia.com/rdp/cudnn-archive" rel="noreferrer">Download cuDNN</a> v7.6.4 for CUDA 10.1. This will require you to sign up to the NVIDIA Developer Program.</li>
<li>Unzip to a suitable location and add the bin directory to your PATH.</li>
</ol></li>
<li>Install tensorflow by <code>pip install tensorflow</code>.</li>
<li>You <a href="https://stackoverflow.com/a/51112550/604687">may need to restart your PC</a>.</li>
</ol>

blocks|key|1611618|text|我用另一种方法解决了这个问题。首先，我从这个链接中安装了Cuda10.1工具包。|type|unstyled|depth|inlineStyleRanges|entityRanges|offset|length|data|1611619|其中我选择了安装程序类型：主管(当地)+(用于windows)和10.1在自定义模式下安装(没有visual+studio集成，NVIDIA+PhysX，因为我以前安装了CUDA+10.2，因此自动安装了所需的依赖项)|1611620|安装之后，我从下面的路径(C:\Program+Files\NVIDIA+GPU+Computing+Toolkit\CUDA\v10.1\bin)复制了cudart64_101.dll文件并粘贴到(C:\Program+Files\NVIDIA+GPU+Computing+Toolkit\CUDA\v10.2\bin)中。|style|CODE|1611621|然后进口了Tensorflow，运行顺利。|1611622|entityMap|0|LINK|mutability|MUTABLE|url|https://developer.nvidia.com/cuda-10.1-download-archive-base|1|https://developer.nvidia.com/cuda-10.1-download-archive-base?target_os=Windows&target_arch=x86_64&target_version=10&target_type=exelocal^0|M|2|0|0|D|6|1|0|D|1O|25|G|2S|1O|0|0^^$0|@$1|2|3|4|5|6|7|V|8|@]|9|@$A|W|B|X|1|Y]]|C|$]]|$1|D|3|E|5|6|7|Z|8|@]|9|@$A|10|B|11|1|12]]|C|$]]|$1|F|3|G|5|6|7|13|8|@$A|14|B|15|H|I]|$A|16|B|17|H|I]|$A|18|B|19|H|I]]|9|@]|C|$]]|$1|J|3|K|5|6|7|1A|8|@]|9|@]|C|$]]|$1|L|3|-4|5|6|7|1B|8|@]|9|@]|C|$]]]|M|$N|$5|O|P|Q|C|$R|S]]|T|$5|O|P|Q|C|$R|U]]]]

I solved this another way.
First of all I installed cuda 10.1 toolkit from this <a href="https://developer.nvidia.com/cuda-10.1-download-archive-base" rel="noreferrer">link</a>.
Where I selected installer type: <a href="https://developer.nvidia.com/cuda-10.1-download-archive-base?target_os=Windows&amp;target_arch=x86_64&amp;target_version=10&amp;target_type=exelocal" rel="noreferrer">exe(local)</a> (for windows) and installed 10.1 in custom mode (without visual studio integration, NVIDIA PhysX because previously I installed CUDA 10.2 so required dependencies were installed automatically)
After installation, From the Following Path
(<code>C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.1\bin</code>)
, in my case, I copied <code>cudart64_101.dll</code> file and pasted in
(<code>C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\bin</code>).
Then importing Tensorflow worked smoothly.

blocks|key|1400886|text|这个解决方案适用于我：|type|unstyled|depth|inlineStyleRanges|entityRanges|offset|length|data|1400887|我用anaconda预装了环境(这是代码)|1400888|conda+create+-n+YOURENVNAME+python=3.6+//+3.6>+incompatible+with+keras
conda+activate+YOURENVNAME
conda+install+tensorflow-gpu
conda+install+-c+anaconda+keras
conda+install+-c+anaconda+scikit-learn
conda+install+matplotlib|code-block|syntax|javascript|1400889|但在我收到这些警告之后|1400890|2020-02-23+13:31:44.910213:+W+tensorflow/stream_executor/platform/default/dso_loader.cc:55]+Could+not+load+dynamic+library+'cudart64_101.dll';+dlerror:+cudart64_101.dll+not+found

2020-02-23+13:31:44.925815:+I+tensorflow/stream_executor/platform/default/dso_loader.cc:44]+Successfully+opened+dynamic+library+cublas64_10.dll

2020-02-23+13:31:44.941384:+I+tensorflow/stream_executor/platform/default/dso_loader.cc:44]+Successfully+opened+dynamic+library+cufft64_10.dll

2020-02-23+13:31:44.947427:+I+tensorflow/stream_executor/platform/default/dso_loader.cc:44]+Successfully+opened+dynamic+library+curand64_10.dll

2020-02-23+13:31:44.965893:+I+tensorflow/stream_executor/platform/default/dso_loader.cc:44]+Successfully+opened+dynamic+library+cusolver64_10.dll

2020-02-23+13:31:44.982990:+I+tensorflow/stream_executor/platform/default/dso_loader.cc:44]+Successfully+opened+dynamic+library+cusparse64_10.dll

2020-02-23+13:31:44.990036:+W+tensorflow/stream_executor/platform/default/dso_loader.cc:55]+Could+not+load+dynamic+library+'cudnn64_7.dll';+dlerror:+cudnn64_7.dll+not+found|1400891|如何解决第一个警告:我只是下载一个包含所有cudnn文件(dll等)的压缩文件：https://developer.nvidia.com/cudnn|1400892|如何解决第二个警告:我查看了由conda创建的虚拟env中最后一个丢失的文件(cudart64_101.dll)，并将其复制/粘贴在与.dll+cudnn相同的lib文件夹中。|1400893|entityMap|0|LINK|mutability|MUTABLE|url|https://www.joe0.com/2019/10/19/how-resolve-tensorflow-2-0-error-could-not-load-dynamic-library-cudart64_100-dll-dlerror-cudart64_100-dll-not-found/|1|https://developer.nvidia.com/cudnn^0|2|4|0|0|0|0|0|0|14|Y|1|0|0^^$0|@$1|2|3|4|5|6|7|12|8|@]|9|@$A|13|B|14|1|15]]|C|$]]|$1|D|3|E|5|6|7|16|8|@]|9|@]|C|$]]|$1|F|3|G|5|H|7|17|8|@]|9|@]|C|$I|J]]|$1|K|3|L|5|6|7|18|8|@]|9|@]|C|$]]|$1|M|3|N|5|H|7|19|8|@]|9|@]|C|$I|J]]|$1|O|3|P|5|6|7|1A|8|@]|9|@$A|1B|B|1C|1|1D]]|C|$]]|$1|Q|3|R|5|6|7|1E|8|@]|9|@]|C|$]]|$1|S|3|-4|5|6|7|1F|8|@]|9|@]|C|$]]]|T|$U|$5|V|W|X|C|$Y|Z]]|10|$5|V|W|X|C|$Y|11]]]]

This <a href="https://www.joe0.com/2019/10/19/how-resolve-tensorflow-2-0-error-could-not-load-dynamic-library-cudart64_100-dll-dlerror-cudart64_100-dll-not-found/" rel="nofollow noreferrer">solution</a> worked for me :
I preinstalled the environnement with anaconda (here is the code)
<pre><code>conda create -n YOURENVNAME python=3.6 // 3.6&gt; incompatible with keras
conda activate YOURENVNAME
conda install tensorflow-gpu
conda install -c anaconda keras
conda install -c anaconda scikit-learn
conda install matplotlib
</code></pre>
but after I had still these warnings
<pre><code>2020-02-23 13:31:44.910213: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'cudart64_101.dll'; dlerror: cudart64_101.dll not found

2020-02-23 13:31:44.925815: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cublas64_10.dll

2020-02-23 13:31:44.941384: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cufft64_10.dll

2020-02-23 13:31:44.947427: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library curand64_10.dll

2020-02-23 13:31:44.965893: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusolver64_10.dll

2020-02-23 13:31:44.982990: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cusparse64_10.dll

2020-02-23 13:31:44.990036: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'cudnn64_7.dll'; dlerror: cudnn64_7.dll not found
</code></pre>
How I solved the first warning :
I just download a zip file wich contained all the cudnn files (dll, etc) here : <a href="https://developer.nvidia.com/cudnn" rel="nofollow noreferrer">https://developer.nvidia.com/cudnn</a>
How I solved the second warning :
I looked the last missing file (cudart64_101.dll) in my virtual env created by conda and I just copy/pasted it in the same lib folder than for the .dll cudnn

blocks|key|1400925|text|一个更简单的方法是创建一个名为cudart64_101.dll的链接来指向cudart64_102.dll。这并不是很正统，但由于TensorFlow正在寻找cudart64_101.dll导出的符号，而nvidia的人不是业余爱好者，他们很可能不会删除101个到102个符号。基于这一假设，它可以工作(里程可能不同)。|type|unstyled|depth|inlineStyleRanges|offset|length|style|CODE|entityRanges|data|1400926|entityMap^0|F|G|11|G|27|G|0^^$0|@$1|2|3|4|5|6|7|H|8|@$9|I|A|J|B|C]|$9|K|A|L|B|C]|$9|M|A|N|B|C]]|D|@]|E|$]]|$1|F|3|-4|5|6|7|O|8|@]|D|@]|E|$]]]|G|$]]

A simpler way would be to create a link called <code>cudart64_101.dll</code> to point to <code>cudart64_102.dll</code>. This is not very orthodox but since TensorFlow is looking for <code>cudart64_101.dll</code> exported symbols and the nvidia folks are not amateurs, they would most likely not remove symbols from 101 to 102. It works, based on this assumption (mileage may vary).

blocks|key|1611749|text|Tensorflow+2.1与Cuda+10.1一起工作。|type|unstyled|depth|inlineStyleRanges|entityRanges|data|1611750|如果你想要一个快速的黑客：|1611751|只需从cudart64_101.dll下载这里。解压缩压缩文件并将cudart64_101.dll复制到CUDA+bin目录|unordered-list-item|offset|length|style|CODE|1611752|其他：|1611753|安装Cuda+10.1|1611754|entityMap|0|LINK|mutability|MUTABLE|url|https://www.dll-files.com/cudart64_101.dll.html^0|0|0|3|G|X|G|1L|3|L|2|0|0|0|0^^$0|@$1|2|3|4|5|6|7|W|8|@]|9|@]|A|$]]|$1|B|3|C|5|6|7|X|8|@]|9|@]|A|$]]|$1|D|3|E|5|F|7|Y|8|@$G|Z|H|10|I|J]|$G|11|H|12|I|J]|$G|13|H|14|I|J]]|9|@$G|15|H|16|1|17]]|A|$]]|$1|K|3|L|5|6|7|18|8|@]|9|@]|A|$]]|$1|M|3|N|5|F|7|19|8|@]|9|@]|A|$]]|$1|O|3|-4|5|6|7|1A|8|@]|9|@]|A|$]]]|P|$Q|$5|R|S|T|A|$U|V]]]]

Tensorflow 2.1 works with Cuda 10.1. 

If you want a quick hack:

<ul>
<li>Just download <code>cudart64_101.dll</code> from <a href="https://www.dll-files.com/cudart64_101.dll.html" rel="nofollow noreferrer">here</a>. Extract the zip file and copy the <code>cudart64_101.dll</code> to your CUDA <code>bin</code> directory</li>
</ul>

Else:

<ul>
<li>Install Cuda 10.1</li>
</ul>

blocks|key|767443|text|在conda环境中，这就是解决问题的原因(我错过了cudart64-100.dll|type|unstyled|depth|inlineStyleRanges|offset|length|style|CODE|entityRanges|data|767444|从100.DLL下载|ordered-list-item|767445|放在我在C:\Users\<user>\Anaconda3\envs\<env+name>\Library\bin的conda环境中|767446|这就够了！你可以再检查一遍它是否有效：|767447|import+tensorflow+as+tf
tf.config.experimental.list_physical_devices('GPU')|code-block|syntax|javascript|767448|entityMap|0|LINK|mutability|MUTABLE|url|https://www.dll-files.com/search/?q=CUDART64_100.DLL^0|1|5|P|G|0|1|7|0|0|4|1H|0|0|0^^$0|@$1|2|3|4|5|6|7|Z|8|@$9|10|A|11|B|C]|$9|12|A|13|B|C]]|D|@]|E|$]]|$1|F|3|G|5|H|7|14|8|@]|D|@$9|15|A|16|1|17]]|E|$]]|$1|I|3|J|5|H|7|18|8|@$9|19|A|1A|B|C]]|D|@]|E|$]]|$1|K|3|L|5|6|7|1B|8|@]|D|@]|E|$]]|$1|M|3|N|5|O|7|1C|8|@]|D|@]|E|$P|Q]]|$1|R|3|-4|5|6|7|1D|8|@]|D|@]|E|$]]]|S|$T|$5|U|V|W|E|$X|Y]]]]

In a <code>conda</code> environment, this is what solved my problem (I was missing <code>cudart64-100.dll</code>:
<ol>
<li>Downloaded it from
<a href="https://www.dll-files.com/search/?q=CUDART64_100.DLL" rel="noreferrer">dll-files.com/CUDART64_100.DLL</a>
</li>
<li>Put it in my conda environment at
<code>C:\Users\&lt;user&gt;\Anaconda3\envs\&lt;env name&gt;\Library\bin</code>
</li>
</ol>
That's all it took! You can double check if it's working:
<pre><code>import tensorflow as tf
tf.config.experimental.list_physical_devices('GPU')
</code></pre>

blocks|key|1611868|text|Tensorflow+gpu+2.2和2.3每晚|type|unstyled|depth|inlineStyleRanges|entityRanges|data|1611869|(沿CUDA工具包11.0+RC)|1611870|为了解决与OP相同的问题，我只需在磁盘上找到cudart64_101.dll+(在我的例子中是C:\Program+\NVIDIA\NvStreamSrv)，并将其添加为可变环境(即向用户的环境变量路径添加值C:\Program+)。|offset|length|style|BOLD|1611871|entityMap^0|0|0|1B|U|2R|2|2W|D|0^^$0|@$1|2|3|4|5|6|7|L|8|@]|9|@]|A|$]]|$1|B|3|C|5|6|7|M|8|@]|9|@]|A|$]]|$1|D|3|E|5|6|7|N|8|@$F|O|G|P|H|I]|$F|Q|G|R|H|I]|$F|S|G|T|H|I]]|9|@]|A|$]]|$1|J|3|-4|5|6|7|U|8|@]|9|@]|A|$]]]|K|$]]

<h2>Tensorflow gpu 2.2 and 2.3 nightly</h2>
(along CUDA Toolkit 11.0 RC)
To solve the same issue as OP, I just had to find cudart64_101.dll on my disk (in my case C:\Program Files\NVIDIA Corporation\NvStreamSrv) and add it as variable environment (that is add value C:\Program Files\NVIDIA\Corporation\NvStreamSrv)cudart64_101.dll to user's environment variable Path).

blocks|key|1401034|text|通过将NVIDIA设备驱动程序更新为最新版本(v446.14)，解决了这个问题。NVIDIA驱动程序下载链接在这里。|type|unstyled|depth|inlineStyleRanges|entityRanges|offset|length|data|1401035|entityMap|0|LINK|mutability|MUTABLE|url|https://www.nvidia.com/download/index.aspx?lang=en-us^0|1G|5|0|0^^$0|@$1|2|3|4|5|6|7|L|8|@]|9|@$A|M|B|N|1|O]]|C|$]]|$1|D|3|-4|5|6|7|P|8|@]|9|@]|C|$]]]|E|$F|$5|G|H|I|C|$J|K]]]]

Was able to fix the issue by updating NVIDIA device drivers to the latest (v446.14).
NVIDIA drivers download <a href="https://www.nvidia.com/download/index.aspx?lang=en-us" rel="nofollow noreferrer">link here</a>.

blocks|key|1401067|text|TensorFlow+2.3.0适用于CUDA+11，但您必须安装tf-nightly(安装了tensorflow和CUDA+11之后)：https://pypi.org/project/tf-nightly-gpu/|type|unstyled|depth|inlineStyleRanges|entityRanges|offset|length|data|1401068|尝试：|1401069|pip+install+tf-nightly-gpu|code-block|syntax|javascript|1401070|之后，您将在控制台中得到消息：|1401071|I+tensorflow/stream_executor/platform/default/dso_loader.cc:48]+Successfully+opened+dynamic+library+cudart64_110.dll|1401072|entityMap|0|LINK|mutability|MUTABLE|url|https://pypi.org/project/tf-nightly-gpu/^0|1X|14|0|0|0|0|0|0^^$0|@$1|2|3|4|5|6|7|W|8|@]|9|@$A|X|B|Y|1|Z]]|C|$]]|$1|D|3|E|5|6|7|10|8|@]|9|@]|C|$]]|$1|F|3|G|5|H|7|11|8|@]|9|@]|C|$I|J]]|$1|K|3|L|5|6|7|12|8|@]|9|@]|C|$]]|$1|M|3|N|5|H|7|13|8|@]|9|@]|C|$I|J]]|$1|O|3|-4|5|6|7|14|8|@]|9|@]|C|$]]]|P|$Q|$5|R|S|T|C|$U|V]]]]

TensorFlow 2.3.0 works fine with CUDA 11. But you have to install tf-nightly-gpu (after you installed tensorflow and CUDA 11):
<a href="https://pypi.org/project/tf-nightly-gpu/" rel="noreferrer">https://pypi.org/project/tf-nightly-gpu/</a>
Try:
<pre><code>pip install tf-nightly-gpu
</code></pre>
Afterwards you'll get the message in your console:
<pre><code>I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library cudart64_110.dll
</code></pre>

blocks|key|1401100|text|如果您看到上面的错误，这个答案可能会有帮助，但是实际上您已经安装了CUDA+10：|type|unstyled|depth|inlineStyleRanges|offset|length|style|BOLD|entityRanges|data|1401101|pip+install+tensorflow-gpu==2.0.0|code-block|syntax|javascript|1401102|产出：|1401103|I+tensorflow/stream_executor/platform/default/dso_loader.cc:44]+Successfully+opened+dynamic+library+cudart64_100.dll|1401104|这是我的解决办法。|1401105|entityMap^0|0|15|0|0|0|0|0^^$0|@$1|2|3|4|5|6|7|S|8|@$9|T|A|U|B|C]]|D|@]|E|$]]|$1|F|3|G|5|H|7|V|8|@]|D|@]|E|$I|J]]|$1|K|3|L|5|6|7|W|8|@]|D|@]|E|$]]|$1|M|3|N|5|H|7|X|8|@]|D|@]|E|$I|J]]|$1|O|3|P|5|6|7|Y|8|@]|D|@]|E|$]]|$1|Q|3|-4|5|6|7|Z|8|@]|D|@]|E|$]]]|R|$]]

This answer might be helpful if you see above error but actually you have CUDA 10 installed:
<pre><code>pip install tensorflow-gpu==2.0.0
</code></pre>
output:
<pre><code>I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_100.dll
</code></pre>
which was the solution for me.

blocks|key|1612179|text|--在我的例子中，tensorflow安装是在寻找cudart64_101.dll|type|unstyled|depth|inlineStyleRanges|offset|length|style|BOLD|entityRanges|data|1612180|​|1612181|📷|atomic|1612182|1612183|cudart64_101的101部分是Cuda版本-这里101+=+10.1|1612184|我下载了11.x，所以我的系统上的cudart64版本是cudart64_110.dll|1612185|1612186|1612187|1612188|，这是错误的文件！！cudart64_101.dll≠cudart64_110.dll|1612189|解决方案|1612190|从https://developer.nvidia.com/下载Cuda+10.1|1612191|安装(我的NSight崩溃，所以我关闭了)|1612192|1612193|1612194|1612195|安装完成后，您应该有一个Cuda10.1文件夹，而在bin中，系统会抱怨缺少dll。|1612196|1612197|1612198|1612199|检查10.1+bin文件夹的路径是否注册为系统环境变量，因此在加载库时将检查该路径|1612200|1612201|1612202|1612203|如果系统没有立即启动路径，可能需要重新启动。|1612204|1612205|1612206|1612207|entityMap|0|IMAGE|mutability|IMMUTABLE|imageUrl|https://developer.qcloudimg.com/http-save/yehe-900000/9eff6f74bef9900e66e60d54ee948771.png|imageAlt|1|https://developer.qcloudimg.com/http-save/yehe-900000/e52bd115f7602914571d0fea6341ab92.png|2|LINK|MUTABLE|url|https://developer.nvidia.com/|3|https://developer.qcloudimg.com/http-save/yehe-900000/25c5b20eb9ecbeccc8f7d13844dfb8b3.png|4|https://developer.qcloudimg.com/http-save/yehe-900000/5cb22bdad30d9e8993fa78cfc213de11.png|5|https://developer.qcloudimg.com/http-save/yehe-900000/9632170733d86e6619e50ed3b4853791.png|6|https://developer.qcloudimg.com/http-save/yehe-900000/b0bcacbdedc057f3987fcdead64aeca6.png^0|0|15|0|0|0|1|0|0|0|0|0|0|0|1|1|0|0|0|17|0|0|1|T|2|0|0|0|0|1|3|0|0|0|0|0|1|4|0|0|0|0|0|1|5|0|0|D|8|0|0|0|1|6|0|0^^$0|@$1|2|3|4|5|6|7|26|8|@$9|27|A|28|B|C]]|D|@]|E|$]]|$1|F|3|G|5|6|7|29|8|@]|D|@]|E|$]]|$1|H|3|I|5|J|7|2A|8|@]|D|@$9|2B|A|2C|1|2D]]|E|$]]|$1|K|3|G|5|6|7|2E|8|@]|D|@]|E|$]]|$1|L|3|M|5|6|7|2F|8|@]|D|@]|E|$]]|$1|N|3|O|5|6|7|2G|8|@]|D|@]|E|$]]|$1|P|3|G|5|6|7|2H|8|@]|D|@]|E|$]]|$1|Q|3|I|5|J|7|2I|8|@]|D|@$9|2J|A|2K|1|2L]]|E|$]]|$1|R|3|G|5|6|7|2M|8|@]|D|@]|E|$]]|$1|S|3|T|5|6|7|2N|8|@$9|2O|A|2P|B|C]]|D|@]|E|$]]|$1|U|3|V|5|6|7|2Q|8|@]|D|@]|E|$]]|$1|W|3|X|5|6|7|2R|8|@]|D|@$9|2S|A|2T|1|2U]]|E|$]]|$1|Y|3|Z|5|6|7|2V|8|@]|D|@]|E|$]]|$1|10|3|G|5|6|7|2W|8|@]|D|@]|E|$]]|$1|11|3|I|5|J|7|2X|8|@]|D|@$9|2Y|A|2Z|1|30]]|E|$]]|$1|12|3|G|5|6|7|31|8|@]|D|@]|E|$]]|$1|13|3|14|5|6|7|32|8|@]|D|@]|E|$]]|$1|15|3|G|5|6|7|33|8|@]|D|@]|E|$]]|$1|16|3|I|5|J|7|34|8|@]|D|@$9|35|A|36|1|37]]|E|$]]|$1|17|3|G|5|6|7|38|8|@]|D|@]|E|$]]|$1|18|3|19|5|6|7|39|8|@]|D|@]|E|$]]|$1|1A|3|G|5|6|7|3A|8|@]|D|@]|E|$]]|$1|1B|3|I|5|J|7|3B|8|@]|D|@$9|3C|A|3D|1|3E]]|E|$]]|$1|1C|3|G|5|6|7|3F|8|@]|D|@]|E|$]]|$1|1D|3|1E|5|6|7|3G|8|@$9|3H|A|3I|B|C]]|D|@]|E|$]]|$1|1F|3|G|5|6|7|3J|8|@]|D|@]|E|$]]|$1|1G|3|I|5|J|7|3K|8|@]|D|@$9|3L|A|3M|1|3N]]|E|$]]|$1|1H|3|G|5|6|7|3O|8|@]|D|@]|E|$]]|$1|1I|3|-4|5|6|7|3P|8|@]|D|@]|E|$]]]|1J|$1K|$5|1L|1M|1N|E|$1O|1P|1Q|-4]]|1R|$5|1L|1M|1N|E|$1O|1S|1Q|-4]]|1T|$5|1U|1M|1V|E|$1W|1X]]|1Y|$5|1L|1M|1N|E|$1O|1Z|1Q|-4]]|20|$5|1L|1M|1N|E|$1O|21|1Q|-4]]|22|$5|1L|1M|1N|E|$1O|23|1Q|-4]]|24|$5|1L|1M|1N|E|$1O|25|1Q|-4]]]]

In my case the tensorflow install was looking for cudart64_101.dll
<a href="https://i.stack.imgur.com/99IEu.png" rel="noreferrer"><img src="https://i.stack.imgur.com/99IEu.png" alt="enter image description here" /></a>
The 101 part of cudart64_101 is the Cuda version - here 101 = 10.1
I had downloaded 11.x, so the version of cudart64 on my system was cudart64_110.dll
<a href="https://i.stack.imgur.com/QNzGY.png" rel="noreferrer"><img src="https://i.stack.imgur.com/QNzGY.png" alt="enter image description here" /></a>
This is the wrong file!! cudart64_101.dll ≠ cudart64_110.dll
<h1>Solution</h1>
Download Cuda 10.1 from <a href="https://developer.nvidia.com/" rel="noreferrer">https://developer.nvidia.com/</a>
Install (mine crashes with NSight Visual Studio Integration, so I switched that off)
<a href="https://i.stack.imgur.com/vxgQf.png" rel="noreferrer"><img src="https://i.stack.imgur.com/vxgQf.png" alt="enter image description here" /></a>
When the install has finished you should have a Cuda 10.1 folder, and in the bin the dll the system was complaining about being missing
<a href="https://i.stack.imgur.com/WWEGf.png" rel="noreferrer"><img src="https://i.stack.imgur.com/WWEGf.png" alt="enter image description here" /></a>
Check that the path to the 10.1 bin folder is registered as a system environmental variable, so it will be checked when loading the library
<a href="https://i.stack.imgur.com/jdTpl.png" rel="noreferrer"><img src="https://i.stack.imgur.com/jdTpl.png" alt="enter image description here" /></a>
You may need a reboot if the path is not picked up by the system straight away
<a href="https://i.stack.imgur.com/Wcpkt.png" rel="noreferrer"><img src="https://i.stack.imgur.com/Wcpkt.png" alt="enter image description here" /></a>

blocks|key|767740|text|我安装了cudatoolkit+11并将dll+C:\Program+Files\NVIDIA+GPU+Computing+Toolkit\CUDA\v11.1\bin复制到C:\Windows\System32。它修正了PyCharm，而不是Anaconda：|type|unstyled|depth|inlineStyleRanges|offset|length|style|CODE|entityRanges|data|767741|名称："/device:CPU:0“device_type："CPU”memory_limit:+268435456局部性{}化身:+6812190123916921346，名称："/device:GPU:0“device_type："GPU”memory_limit:+13429637120+locality+{+bus_id:+1
链接{}+physical_device_desc:+18025633343883307728+P5000：“设备:+0，名称:+Quadro+P5000，pci总线id:+0000:02:00.0，计算能力:+6.1”|blockquote|767742|entityMap^0|O|1O|2F|J|0|0^^$0|@$1|2|3|4|5|6|7|K|8|@$9|L|A|M|B|C]|$9|N|A|O|B|C]]|D|@]|E|$]]|$1|F|3|G|5|H|7|P|8|@]|D|@]|E|$]]|$1|I|3|-4|5|6|7|Q|8|@]|D|@]|E|$]]]|J|$]]

I installed cudatoolkit 11 and copy dll
<code>C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.1\bin</code> to <code>C:\Windows\System32</code>.
It fixed for PyCharm but not for Anaconda jupyter:
<blockquote>
[name: &quot;/device:CPU:0&quot; device_type: &quot;CPU&quot; memory_limit: 268435456
locality { } incarnation: 6812190123916921346 , name: &quot;/device:GPU:0&quot;
device_type: &quot;GPU&quot; memory_limit: 13429637120 locality { bus_id: 1 
links { } } incarnation: 18025633343883307728 physical_device_desc:
&quot;device: 0, name: Quadro P5000, pci bus id: 0000:02:00.0, compute
capability: 6.1&quot; ]
</blockquote>

blocks|key|767790|text|下载CUDA工具包11.0+RC|type|unstyled|depth|inlineStyleRanges|entityRanges|data|767791|为了解决这个问题，我只是在我的磁盘上找到了cudart64_101.dll+(C:\Program+\NVIDIA\NvStreamSrv)，并将其添加为添加值的可变环境(+C:\Program到用户的环境变量路径)。|767792|entityMap^0|0|0^^$0|@$1|2|3|4|5|6|7|F|8|@]|9|@]|A|$]]|$1|B|3|C|5|6|7|G|8|@]|9|@]|A|$]]|$1|D|3|-4|5|6|7|H|8|@]|9|@]|A|$]]]|E|$]]

download CUDA Toolkit 11.0 RC
To solve the issue,
I just find cudart64_101.dll on my disk
( C:\Program Files\NVIDIA Corporation\NvStreamSrv) and add it as variable environment that is add value (C:\Program Files\NVIDIA\Corporation\NvStreamSrv)cudart64_101.dll to user's environment variable Path).

blocks|key|1401447|text|我在混合pip+&+conda以安装tensorflow+2.3时遇到了这个问题。(当conda的tensorflow+2.3安装中断时，我使用pip安装tensorflow+2.3b/c。)|type|unstyled|depth|inlineStyleRanges|entityRanges|data|1401448|最后，我安装了不正确的cudatoolkit和cudnn版本。|1401449|为了解决这个问题，我只使用指定了cudatoolkit和cuda的特定版本的conda+install。|offset|length|style|CODE|1401450|查看配置获得关于tensorflow、cudatoolkit和cuda版本的信息，这些版本应该一起工作。|1401451|entityMap|0|LINK|mutability|MUTABLE|url|https://www.tensorflow.org/install/source_windows?force_isolation=true#tested_build_configurations^0|0|0|12|D|0|2|2|0|0^^$0|@$1|2|3|4|5|6|7|T|8|@]|9|@]|A|$]]|$1|B|3|C|5|6|7|U|8|@]|9|@]|A|$]]|$1|D|3|E|5|6|7|V|8|@$F|W|G|X|H|I]]|9|@]|A|$]]|$1|J|3|K|5|6|7|Y|8|@]|9|@$F|Z|G|10|1|11]]|A|$]]|$1|L|3|-4|5|6|7|12|8|@]|9|@]|A|$]]]|M|$N|$5|O|P|Q|A|$R|S]]]]

I ran into this problem when mixing pip &amp; conda to get tensorflow 2.3 installed. (I used pip to install tensorflow 2.3 b/c at the time conda's install of tensorflow 2.3 was broken.)
I ended up with the incorrect versions of cudatoolkit and cudnn installed.
To solve the problem, I simply did <code>conda install</code> with specific versions of cudatoolkit and cuda specified.
Look at <a href="https://www.tensorflow.org/install/source_windows?force_isolation=true#tested_build_configurations" rel="nofollow noreferrer">https://www.tensorflow.org/install/source_windows?force_isolation=true#tested_build_configurations</a> for info on tensorflow, cudatoolkit, and cuda versions that should work together.

blocks|key|1612376|text|这可能是由您正在运行的python版本引起的，我正在使用microsoft商店的python+3.7，我遇到了这个错误，切换到python+3.10修复了它。|type|unstyled|depth|inlineStyleRanges|entityRanges|data|1612377|entityMap^0|0^^$0|@$1|2|3|4|5|6|7|D|8|@]|9|@]|A|$]]|$1|B|3|-4|5|6|7|E|8|@]|9|@]|A|$]]]|C|$]]

This could be caused by the version of python you are running as well, I was using the python 3.7 from the microsoft store and I run into this error, switching to python 3.10 fixed it.

blocks|key|1401544|text|我在这里遇到了这个错误，在我更新GPU驱动程序之后，问题是我从cmd运行了tensorboard，在这里我没有安装任何when+you+install+TensorFlow+using+anaconda+all+the+required+cuda+and+cudnn+are+downloaded+for+you，因为我使用anaconda+TensorFlow，如果没有使用anaconda+env安装TensorFlow，就会错过文件。|type|unstyled|depth|inlineStyleRanges|offset|length|style|CODE|entityRanges|data|1401545|​|1401546|📷|atomic|1401547|1401548|解决方案|1401549|*只要打开龙头板就行了|1401550|1-或者只是下载最后一个Cuda工具包并将其添加到路径中。|1401551|https://developer.nvidia.com/cuda-toolkit-archive|1401552|使用这个2-conda+install+-c+anaconda+cudatoolkit|1401553|3-然后重新启动你的电脑|1401554|entityMap|0|IMAGE|mutability|IMMUTABLE|imageUrl|https://developer.qcloudimg.com/http-save/yehe-900000/c13bd12cf6c879fc6f53c56c21807f7b.png|imageAlt|1|LINK|MUTABLE|url^0|1N|2P|0|0|0|1|0|0|0|0|0|0|0|1D|1|0|6|11|0|0^^$0|@$1|2|3|4|5|6|7|1A|8|@$9|1B|A|1C|B|C]]|D|@]|E|$]]|$1|F|3|G|5|6|7|1D|8|@]|D|@]|E|$]]|$1|H|3|I|5|J|7|1E|8|@]|D|@$9|1F|A|1G|1|1H]]|E|$]]|$1|K|3|G|5|6|7|1I|8|@]|D|@]|E|$]]|$1|L|3|M|5|6|7|1J|8|@]|D|@]|E|$]]|$1|N|3|O|5|6|7|1K|8|@]|D|@]|E|$]]|$1|P|3|Q|5|6|7|1L|8|@]|D|@]|E|$]]|$1|R|3|S|5|6|7|1M|8|@]|D|@$9|1N|A|1O|1|1P]]|E|$]]|$1|T|3|U|5|6|7|1Q|8|@$9|1R|A|1S|B|C]]|D|@]|E|$]]|$1|V|3|W|5|6|7|1T|8|@]|D|@]|E|$]]|$1|X|3|-4|5|6|7|1U|8|@]|D|@]|E|$]]]|Y|$Z|$5|10|11|12|E|$13|14|15|-4]]|16|$5|17|11|18|E|$19|S]]]]

I had this error here with tensorboard , it happen after I update the GPU driver
the thing is i was running tensorboard from the cmd where I didn't install any CUDA since I was using anaconda
<code>when you install TensorFlow using anaconda all the required cuda and cudnn are downloaded for you </code> you will miss the files if you didn't use anaconda env where you install TensorFlow in it
<a href="https://i.stack.imgur.com/NKJRF.png" rel="nofollow noreferrer"><img src="https://i.stack.imgur.com/NKJRF.png" alt="enter image description here" /></a>
<h2>Solution</h2>
<h3>* just open tensoboard from anaconda</h3>
<h3>1-or just download the last Cuda toolkit and add it to paths</h3>
<a href="https://developer.nvidia.com/cuda-toolkit-archive" rel="nofollow noreferrer">https://developer.nvidia.com/cuda-toolkit-archive</a>
and use this
2-<code>conda install -c anaconda cudatoolkit</code>
3-then restart your pc

blocks|key|1612575|text|我也面临过类似的问题，并意识到问题在于数据自动化系统和CUDNN版本不匹配。|type|unstyled|depth|inlineStyleRanges|entityRanges|data|1612576|可以参考这里获得正确的版本。在下面参考的TensorFlow+2.4.0中，建议使用CUDA+11.0和cuDNN+8.0。|offset|length|1612577|📷|atomic|1612578|​|1612579|或者你可以参考这里下载cuDNN以获得合适的数据自动化系统。|1612580|1612581|1612582|entityMap|0|LINK|mutability|MUTABLE|url|https://www.tensorflow.org/install/source_windows|1|IMAGE|IMMUTABLE|imageUrl|https://developer.qcloudimg.com/http-save/yehe-900000/09982d552dd21a58b42f42dfc9c475b8.png|imageAlt|2|https://developer.nvidia.com/rdp/cudnn-archive|3|https://developer.qcloudimg.com/http-save/yehe-900000/dd31714e03044b3d2037c70a86e89b4f.png^0|0|4|2|0|0|0|1|1|0|0|7|2|2|0|0|1|3|0|0^^$0|@$1|2|3|4|5|6|7|16|8|@]|9|@]|A|$]]|$1|B|3|C|5|6|7|17|8|@]|9|@$D|18|E|19|1|1A]]|A|$]]|$1|F|3|G|5|H|7|1B|8|@]|9|@$D|1C|E|1D|1|1E]]|A|$]]|$1|I|3|J|5|6|7|1F|8|@]|9|@]|A|$]]|$1|K|3|L|5|6|7|1G|8|@]|9|@$D|1H|E|1I|1|1J]]|A|$]]|$1|M|3|G|5|H|7|1K|8|@]|9|@$D|1L|E|1M|1|1N]]|A|$]]|$1|N|3|J|5|6|7|1O|8|@]|9|@]|A|$]]|$1|O|3|-4|5|6|7|1P|8|@]|9|@]|A|$]]]|P|$Q|$5|R|S|T|A|$U|V]]|W|$5|X|S|Y|A|$Z|10|11|-4]]|12|$5|R|S|T|A|$U|13]]|14|$5|X|S|Y|A|$Z|15|11|-4]]]]

I too have faced similar issues and realized that the issue was with CUDA and CUDNN version mismatch.
Can refer <a href="https://www.tensorflow.org/install/source_windows" rel="nofollow noreferrer">here</a> for the proper versions. From the reference below for TensorFlow 2.4.0 it is recommended to use CUDA 11.0 and cuDNN 8.0.
<a href="https://i.stack.imgur.com/AfnmD.png" rel="nofollow noreferrer"><img src="https://i.stack.imgur.com/AfnmD.png" alt="CUDA &amp; CUDNN reference for TensorFlow" /></a>
Or you can refer <a href="https://developer.nvidia.com/rdp/cudnn-archive" rel="nofollow noreferrer">here</a> to download cuDNN for suitable CUDA.
<a href="https://i.stack.imgur.com/jLTID.png" rel="nofollow noreferrer"><img src="https://i.stack.imgur.com/jLTID.png" alt="cuDNN for CUDA" /></a>

I just installed the latest version of Tensorflow via <code>pip install tensorflow</code> and whenever I run a program, I get the log message:

<blockquote>
 W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'cudart64_101.dll'; dlerror: cudart64_101.dll not found
</blockquote>

Is this bad? How do I fix the error?

Could not load dynamic library 'cudart64_101.dll' on tensorflow CPU-only installation

翻译质量差，导致语言生硬或混乱。

没有提供实际的解决方法或示例。

解答不清晰，无法理解或解决问题。

页面排版不美观，阅读体验差。

文章

问答

视频

学习中心

腾讯云实验室

直播

竞赛

腾讯云代码分析专区

腾讯iOA零信任安全管理系统专区

腾讯云架构师技术同盟交流圈

腾讯云数据库专区

腾讯云顾问专区

腾讯云原生专区

腾讯混元专区

腾讯云TCE专区

腾讯云Lighthouse专区

腾讯云HAI专区

腾讯云Edgeone专区

腾讯云存储专区

腾讯云智能专区

腾讯轻联专区 

腾讯云开发专区

TAPD专区

腾讯轻量云游戏服专区

腾讯云最具价值专家

腾讯云架构师技术同盟

腾讯云创作之星

腾讯云开发者先锋 

腾讯云代码助手

CODING DevOps

Cloud Studio

SDK中心

API中心

命令行工具

我刚刚通过pip install tensorflow安装了Tensorflow的最新版本，每当我运行一个程序时，我都会收到日志消息：W tensorflow/stream_executor/platform/default/dso_loader.cc:55]无法加载动态库cudart64 64_101.dll；dle...

问无法在tensorflow CPU安装上加载动态库“cudart64_101.dll”
EN

Stack Overflow用户

社区

活动

圈层

关于

腾讯云开发者

热门产品

热门推荐

更多推荐

问无法在tensorflow CPU安装上加载动态库“cudart64_101.dll”EN