Unable to Run Tensorflow/Keras with GPU

Question

I tried to run Keras with my GPU but got the following error:

C:\Python36\lib\site-packages\skimage\transform_warps.py:84: UserWarning: The default mode, 'constant', will be changed to 'reflect' in skimage 0.15. warn("The default mode, 'constant', will be changed to 'reflect' in " E C:\tf_jenkins\workspace\rel-win\M\windows-gpu\PY\36\tensorflow\stream_executor\cuda\cuda_dnn.cc:378]

Loaded runtime CuDNN library: 7102 (compatibility version 7100) but source was compiled with 7003 (compatibility version 7000). If using a binary install, upgrade your CuDNN library to match. If building from sources, make sure the library loaded at runtime matches a compatible version specified during compile configuration.

F C:\tf_jenkins\workspace\rel-win\M\windows-gpu\PY\36\tensorflow\core\kernels\conv_ops.cc:717] Check failed: stream->parent()->GetConvolveAlgorithms( conv_parameters.ShouldIncludeWinogradNonfusedAlgo(), &algorithms)

I have tensorflow 1.6, CUDA version: Cuda compilation tools, release 9.0, V9.0.176

Does anyone know whats wrong here?

I just downloaded the right drivers but i still get the error. Is there a way to check if i have the correct version of CUDNN? — user2505650
– user2505650, Commented Mar 26, 2018 at 10:53
yes there is and you have already done it. The error message tellsmyu exactly which version of CuDNN you have installed and which version you requre.... — talonmies
– talonmies, Commented Mar 26, 2018 at 11:03

Rookie · Accepted Answer · 2018-03-26 17:59:51Z

8

You need to install cuDNN 7.0.5. The file can be downloaded here. After clicking Download and agreeing to the terms, the option will be listed.

edited Mar 26, 2018 at 17:59

answered Mar 26, 2018 at 16:58

Rookie

3052 silver badges11 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

AllBecomesGood Over a year ago

My I ask how to find out which version is needed? (Like, how do you know he needs 7.0.5). I downloaded cuda 9.0, then afterwards cudnn 7.1.2 for cuda 9.0, so it is odd to me why it is not compatible.

Rookie Over a year ago

Notice that the error received reads: Loaded runtime CuDNN library: 7102 (compatibility version 7100) but source was compiled with 7003 (compatibility version 7000). This indicates the user had installed cuDNN 7.1 (7100) version but the source library was compiled with cuDNN 7.0 (7000) which led to incompatibility. The latest version of the 7000 series is cuDNN 7.0.5.

Collectives™ on Stack Overflow

Unable to Run Tensorflow/Keras with GPU

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related