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

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

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

Is this bad? How do I fix the error?

This question is related to python python-3.x tensorflow keras tensorflow2.0

The answer is


Tensorflow 2.1+

What's going on?

With the new Tensorflow 2.1 release, the default tensorflow 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):

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.

Should I worry? How do I fix it?

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 adapt TF's logging level to suppress warnings, but that might be overkill, as it will silence all warnings.


Tensorflow 1.X or 2.0:

Your CUDA setup is broken, ensure you have the correct version installed.


To install the prerequisites for GPU support in TensorFlow 2.1:

  1. Install your latest GPU drivers.
  2. Install CUDA 10.1.
    • 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.
    • 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.
  3. Install cuDNN.
    1. Download cuDNN v7.6.4 for CUDA 10.1. This will require you to sign up to the NVIDIA Developer Program.
    2. Unzip to a suitable location and add the bin directory to your PATH.
  4. Install tensorflow by pip install tensorflow.
  5. You may need to restart your PC.

TensorFlow 2.3.0 works fine with CUDA 11. But you have to install tf-nightly-gpu (after you installed tensorflow and CUDA 11): https://pypi.org/project/tf-nightly-gpu/

Try:

pip install tf-nightly-gpu

Afterwards you'll get the message in your console:

I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library cudart64_110.dll

I solved this another way. First of all I installed cuda 10.1 toolkit from this link

Where i selected installer type(exe(local)) and installed 10.1 in custom mode means (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 (C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.1\bin) , in my case, I copied 'cudart64_101.dll' file and pasted in (C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.2\bin).

Then importing Tensorflow worked smoothly.

N.B. Sorry for Bad English


In a conda environment, this is what solved my problem (I was missing cudart64-100.dll:

  1. Downloaded it from dll-files.com/CUDART64_100.DLL

  2. Put it in my conda environment at C:\Users\<user>\Anaconda3\envs\<env name>\Library\bin

That's all it took! You can double check if it's working:

import tensorflow as tf
tf.config.experimental.list_physical_devices('GPU')

In my case the tensorflow install was looking for cudart64_101.dll

enter image description here

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

enter image description here

This is the wrong file!! cudart64_101.dll ? cudart64_110.dll

Solution

Download Cuda 10.1 from https://developer.nvidia.com/

Install (mine crashes with NSight Visual Studio Integration, so I switched that off)

enter image description here

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

enter image description here

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

enter image description here

You may need a reboot if the path is not picked up by the system straight away

enter image description here


This solution worked for me :

I preinstalled the environnement with anaconda (here is the code)

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

but after I had still these warnings

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

How I solved the first warning : I just download a zip file wich contained all the cudnn files (dll, etc) here : https://developer.nvidia.com/cudnn

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


Tensorflow 2.1 works with Cuda 10.1.

If you want a quick hack:

  • Just download cudart64_101.dll from here. Extract the zip file and copy the cudart64_101.dll to your CUDA bin directory

Else:

  • Install Cuda 10.1

I installed cudatoolkit 11 and copy dll C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.1\bin to C:\Windows\System32. It fixed for PyCharm but not for Anaconda jupyter:

[name: "/device:CPU:0" device_type: "CPU" memory_limit: 268435456 locality { } incarnation: 6812190123916921346 , name: "/device:GPU:0" device_type: "GPU" memory_limit: 13429637120 locality { bus_id: 1
links { } } incarnation: 18025633343883307728 physical_device_desc: "device: 0, name: Quadro P5000, pci bus id: 0000:02:00.0, compute capability: 6.1" ]


This answer might be helpful if you see above error but actually you have CUDA 10 installed:

pip install tensorflow-gpu==2.0.0

output:

I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudart64_100.dll

which was the solution for me.


Tensorflow gpu 2.2 and 2.3 nightly

(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).


Was able to fix the issue by updating NVIDIA device drivers to the latest (v446.14). NVIDIA drivers download link here.


A simpler way would be to create a link called cudart64_101.dll to point to cudart64_102.dll. This is not very orthodox but since TensorFlow is looking for cudart64_101.dll 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).


Questions with python tag:

programming a servo thru a barometer Is there a way to view two blocks of code from the same file simultaneously in Sublime Text? python variable NameError Why my regexp for hyphenated words doesn't work? Comparing a variable with a string python not working when redirecting from bash script is it possible to add colors to python output? Get Public URL for File - Google Cloud Storage - App Engine (Python) Real time face detection OpenCV, Python xlrd.biffh.XLRDError: Excel xlsx file; not supported Could not load dynamic library 'cudart64_101.dll' on tensorflow CPU-only installation Upgrade to python 3.8 using conda Unable to allocate array with shape and data type How to fix error "ERROR: Command errored out with exit status 1: python." when trying to install django-heroku using pip How to prevent Google Colab from disconnecting? "UserWarning: Matplotlib is currently using agg, which is a non-GUI backend, so cannot show the figure." when plotting figure with pyplot on Pycharm How to fix 'Object arrays cannot be loaded when allow_pickle=False' for imdb.load_data() function? "E: Unable to locate package python-pip" on Ubuntu 18.04 Tensorflow 2.0 - AttributeError: module 'tensorflow' has no attribute 'Session' Jupyter Notebook not saving: '_xsrf' argument missing from post How to Install pip for python 3.7 on Ubuntu 18? Python: 'ModuleNotFoundError' when trying to import module from imported package OpenCV TypeError: Expected cv::UMat for argument 'src' - What is this? Requests (Caused by SSLError("Can't connect to HTTPS URL because the SSL module is not available.") Error in PyCharm requesting website How to setup virtual environment for Python in VS Code? Pylint "unresolved import" error in Visual Studio Code Pandas Merging 101 Numpy, multiply array with scalar What is the meaning of "Failed building wheel for X" in pip install? Selenium: WebDriverException:Chrome failed to start: crashed as google-chrome is no longer running so ChromeDriver is assuming that Chrome has crashed Could not install packages due to an EnvironmentError: [Errno 13] OpenCV !_src.empty() in function 'cvtColor' error ConvergenceWarning: Liblinear failed to converge, increase the number of iterations How to downgrade python from 3.7 to 3.6 I can't install pyaudio on Windows? How to solve "error: Microsoft Visual C++ 14.0 is required."? Iterating over arrays in Python 3 How do I install opencv using pip? How do I install Python packages in Google's Colab? How do I use TensorFlow GPU? How to upgrade Python version to 3.7? How to resolve TypeError: can only concatenate str (not "int") to str How can I install a previous version of Python 3 in macOS using homebrew? Flask at first run: Do not use the development server in a production environment TypeError: only integer scalar arrays can be converted to a scalar index with 1D numpy indices array What is the difference between Jupyter Notebook and JupyterLab? Pytesseract : "TesseractNotFound Error: tesseract is not installed or it's not in your path", how do I fix this? Could not install packages due to a "Environment error :[error 13]: permission denied : 'usr/local/bin/f2py'" How do I resolve a TesseractNotFoundError? Trying to merge 2 dataframes but get ValueError Authentication plugin 'caching_sha2_password' is not supported Python Pandas User Warning: Sorting because non-concatenation axis is not aligned

Questions with python-3.x tag:

Could not load dynamic library 'cudart64_101.dll' on tensorflow CPU-only installation Replace specific text with a redacted version using Python Upgrade to python 3.8 using conda "Permission Denied" trying to run Python on Windows 10 Python: 'ModuleNotFoundError' when trying to import module from imported package What is the meaning of "Failed building wheel for X" in pip install? How to downgrade python from 3.7 to 3.6 I can't install pyaudio on Windows? How to solve "error: Microsoft Visual C++ 14.0 is required."? Iterating over arrays in Python 3 How to upgrade Python version to 3.7? TypeError: only integer scalar arrays can be converted to a scalar index with 1D numpy indices array How do I resolve a TesseractNotFoundError? Could not find a version that satisfies the requirement tensorflow Not able to pip install pickle in python 3.6 json.decoder.JSONDecodeError: Extra data: line 2 column 1 (char 190) installing urllib in Python3.6 pip install returning invalid syntax Unable to import path from django.urls Display all dataframe columns in a Jupyter Python Notebook How to make Firefox headless programmatically in Selenium with Python? How to import cv2 in python3? Pipenv: Command Not Found Error in Python script "Expected 2D array, got 1D array instead:"? Fixed digits after decimal with f-strings How do I upgrade the Python installation in Windows 10? Pip error: Microsoft Visual C++ 14.0 is required Python error message io.UnsupportedOperation: not readable Anaconda Installed but Cannot Launch Navigator Conda command is not recognized on Windows 10 TypeError: can't pickle _thread.lock objects How do you fix the "element not interactable" exception? How to print a specific row of a pandas DataFrame? Relative imports - ModuleNotFoundError: No module named x SyntaxError: unexpected EOF while parsing ImportError: No module named 'django.core.urlresolvers' Why Python 3.6.1 throws AttributeError: module 'enum' has no attribute 'IntFlag'? What is the purpose of "pip install --user ..."? Add Legend to Seaborn point plot How to install pip for Python 3.6 on Ubuntu 16.10? Python sockets error TypeError: a bytes-like object is required, not 'str' with send function WinError 2 The system cannot find the file specified (Python) Python 3.6 install win32api? error UnicodeDecodeError: 'utf-8' codec can't decode byte 0xff in position 0: invalid start byte Python 3 - ValueError: not enough values to unpack (expected 3, got 2) matplotlib: plot multiple columns of pandas data frame on the bar chart Unable to set default python version to python3 in ubuntu TypeError: '<=' not supported between instances of 'str' and 'int' pandas: merge (join) two data frames on multiple columns Replacing a character from a certain index Scrolling to element using webdriver?

Questions with tensorflow tag:

Could not load dynamic library 'cudart64_101.dll' on tensorflow CPU-only installation Module 'tensorflow' has no attribute 'contrib' Tensorflow 2.0 - AttributeError: module 'tensorflow' has no attribute 'Session' Could not install packages due to an EnvironmentError: [WinError 5] Access is denied: How do I use TensorFlow GPU? Which TensorFlow and CUDA version combinations are compatible? Could not find a version that satisfies the requirement tensorflow pip3: command not found How to import keras from tf.keras in Tensorflow? Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX AVX2 Tensorflow import error: No module named 'tensorflow' How to downgrade tensorflow, multiple versions possible? Can I run Keras model on gpu? How to check if keras tensorflow backend is GPU or CPU version? how to install tensorflow on anaconda python 3.6 What is the role of "Flatten" in Keras? Running Tensorflow in Jupyter Notebook Update TensorFlow Tensorflow r1.0 : could not a find a version that satisfies the requirement tensorflow ImportError: No module named tensorflow On Windows, running "import tensorflow" generates No module named "_pywrap_tensorflow" error Keras, How to get the output of each layer? What is the meaning of the word logits in TensorFlow? How to compile Tensorflow with SSE4.2 and AVX instructions? Installing TensorFlow on Windows (Python 3.6.x) Can Keras with Tensorflow backend be forced to use CPU or GPU at will? How to get Tensorflow tensor dimensions (shape) as int values? TensorFlow ValueError: Cannot feed value of shape (64, 64, 3) for Tensor u'Placeholder:0', which has shape '(?, 64, 64, 3)' Deep-Learning Nan loss reasons Cannot import keras after installation TensorFlow not found using pip How to get current available GPUs in tensorflow? How to find which version of TensorFlow is installed in my system? How to tell if tensorflow is using gpu acceleration from inside python shell? How can I run Tensorboard on a remote server? Tensorflow set CUDA_VISIBLE_DEVICES within jupyter Using Keras & Tensorflow with AMD GPU What is the difference between 'SAME' and 'VALID' padding in tf.nn.max_pool of tensorflow? How to run Tensorflow on CPU TensorFlow, "'module' object has no attribute 'placeholder'" Installing tensorflow with anaconda in windows How to get the dimensions of a tensor (in TensorFlow) at graph construction time? How to convert numpy arrays to standard TensorFlow format? In Tensorflow, get the names of all the Tensors in a graph TensorFlow: "Attempting to use uninitialized value" in variable initialization Disable Tensorflow debugging information What is logits, softmax and softmax_cross_entropy_with_logits? How to assign a value to a TensorFlow variable? How to prevent tensorflow from allocating the totality of a GPU memory? Convert a tensor to numpy array in Tensorflow?

Questions with keras tag:

Could not load dynamic library 'cudart64_101.dll' on tensorflow CPU-only installation How to fix 'Object arrays cannot be loaded when allow_pickle=False' for imdb.load_data() function? Tensorflow 2.0 - AttributeError: module 'tensorflow' has no attribute 'Session' What is the use of verbose in Keras while validating the model? Save and load weights in keras How to import keras from tf.keras in Tensorflow? How to check which version of Keras is installed? Can I run Keras model on gpu? How to check if keras tensorflow backend is GPU or CPU version? Keras input explanation: input_shape, units, batch_size, dim, etc How to predict input image using trained model in Keras? What is the role of "Flatten" in Keras? Running Tensorflow in Jupyter Notebook How to concatenate two layers in keras? Why plt.imshow() doesn't display the image? How to save final model using keras? How do I use the Tensorboard callback of Keras? Why binary_crossentropy and categorical_crossentropy give different performances for the same problem? Keras, How to get the output of each layer? Error when checking model input: expected convolution2d_input_1 to have 4 dimensions, but got array with shape (32, 32, 3) Can Keras with Tensorflow backend be forced to use CPU or GPU at will? Deep-Learning Nan loss reasons Cannot import keras after installation Get class labels from Keras functional model Using Keras & Tensorflow with AMD GPU Keras, how do I predict after I trained a model? How to return history of validation loss in Keras Keras model.summary() result - Understanding the # of Parameters How to load a model from an HDF5 file in Keras? Where do I call the BatchNormalization function in Keras? How do I install Keras and Theano in Anaconda Python on Windows?

Questions with tensorflow2.0 tag:

Could not load dynamic library 'cudart64_101.dll' on tensorflow CPU-only installation Tensorflow 2.0 - AttributeError: module 'tensorflow' has no attribute 'Session' How to prevent tensorflow from allocating the totality of a GPU memory?