Downgrading CUDA 10.1 to 10 in windows because TF2.0.0 doesnt work with 10.1

Accelerated Computing CUDA CUDA Setup and Installation
1

I installed CUDA 10.1 and then found out that tensorflow doesnt work with it.Is there any way to downgrade CUDA 10.1 on windows 10 without having to uninstall the whole thing and download it again?

1 Like
2

I think a have the same issue. Does tf also look for an outdated dll?

W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library ‘cudart64_100.dll’; dlerror: cudart64_100.dll not found

3

I think a have the same issue. Does tf also look for an outdated dll?

W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library ‘cudart64_100.dll’; dlerror: cudart64_100.dll not found

4

I also have the same issue. Where can one find the install for 10 rather than 10.1?

I’m perfectly willing to uninstall and reinstall.

5

I found CUDA 10 under “Legacy” at developer.nvidia.com. However, my attempt to install (after uninstall of 10.1) got stuck at this window:

[Album] Inactive NVIDIA CUDA Install Window

Although there is an apparent “Install” option, none of the options respond to mouse clicks.

6

After aborting the install, I tried it again, and it succeeded.

(base) PS C:\Users\jim_b> nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2018 NVIDIA Corporation
Built on Sat_Aug_25_21:08:04_Central_Daylight_Time_2018
Cuda compilation tools, release 10.0, V10.0.130

7

UPDATE: I didn’t have tensorflow-gpu installed – only tensorflow. doing a conda install tensorflow-gpu resolved my particular problem.

However, even after all that, tensorflow doesn’t recognize my dual 1070 GPUs:

(tensorflow) PS C:\Users\jim_b> ipython
Python 3.6.9 |Anaconda, Inc.| (default, Jul 30 2019, 14:00:49) [MSC v.1915 64 bit (AMD64)]
Type ‘copyright’, ‘credits’ or ‘license’ for more information
IPython 7.10.1 – An enhanced Interactive Python. Type ‘?’ for help.

In [1]: import tensorflow as tf

In [2]: tf.version
Out[2]: ‘2.0.0’

In [3]: tf.test.is_gpu_available()
2019-12-09 16:50:55.953081: I tensorflow/core/platform/cpu_feature_guard.cc:145] This TensorFlow binary is optimized with Intel(R) MKL-DNN to use the following CPU instructions in performance critical operations: AVX AVX2
To enable them in non-MKL-DNN operations, rebuild TensorFlow with the appropriate compiler flags.
Out[3]: False

8

CUDA 10.2 works with Tensorflow 2.0, sort of.

By “sort of” I mean I can run a keras sequential model and it reports it is training for a while and then it freezes.

It is unclear how one:

  1. sets verbose logging to output debugging information adequate to track down a freeze like this CTRL-C doesn’t provide much information)

  2. gathers all version information to include in a bug report.

9

Having similar issue

I installed Cuda 10.2 with the associated cudnn with no issues. Installed Tensorflow 2.0. It doesn’t look its recognizing the GPU.

I know with Cuda 10, I won’t have an issue with Tensorflow 2.0

The big question, I am trying to answer. Does Tensorflow 2.0 work with Cuda 10.2.

>>> import tensorflow as tf
>>>  tf.test.is_gpu_available()
  File "<stdin>", line 1
    tf.test.is_gpu_available()
    ^
IndentationError: unexpected indent
>>>

tarik-deeplearning@tarikdeeplearning-All-Series:~$ nvidia-smi
Sat Dec 28 21:21:26 2019       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 440.44       Driver Version: 440.44       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  TITAN X (Pascal)    Off  | 00000000:01:00.0  On |                  N/A |
| 23%   39C    P8    15W / 250W |    275MiB / 12194MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+
|   1  TITAN Xp            Off  | 00000000:02:00.0 Off |                  N/A |
| 23%   30C    P8     9W / 250W |      2MiB / 12196MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID   Type   Process name                             Usage      |
|=============================================================================|
|    0      1338      G   /usr/lib/xorg/Xorg                            18MiB |
|    0      1414      G   /usr/bin/gnome-shell                          49MiB |
|    0      1622      G   /usr/lib/xorg/Xorg                           108MiB |
|    0      1766      G   /usr/bin/gnome-shell                          95MiB |
+-----------------------------------------------------------------------------+

kdeeplearning-All-Series:~$ nvcc -V
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
10

Cuda 10.2 supports tensorflow 1.15. You can try downgrading the version to tensorflow 1.15. My python version is 3.6.8.