I am trying to install tensorflow on my desktop but am having issues.
I am using:
- Windows 10 home edition
- python 3.10.9
- tensorflow 2.10.0
- cudatoolkit 11.2.2
- cudnn 8.1.0.77
- NVIDIA GeForce GTX 1080
- Driver version 516.94
I followed this guide with the only difference that I used Anaconda instead of miniconda.
I also followed this video to install the CudaToolKit and cudNN libraries, and copied over the CUDNN libs into the cuda tool kit folder
The tensorflow installation guide mentions 2 verification commands.
For the first one i get this output:
>python -c "import tensorflow as tf; print(tf.reduce_sum(tf.random.normal([1000, 1000])))"
2023-06-13 20:15:11.453525: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2023-06-13 20:15:11.456435: I tensorflow/core/common_runtime/process_util.cc:146] Creating new thread pool with default inter op setting: 2. Tune using inter_op_parallelism_threads for best performance.
tf.Tensor(-1362.5055, shape=(), dtype=float32)
For the second one I get this output:
>python -c "import tensorflow as tf; print(tf.config.list_physical_devices('GPU'))"
[]
it seems like tensorflow is installed, but it cannot find my GPU. Any pointers as to what could have gone wrong?
1 Like
@hashbang,
Welcome to the Tensorflow Forum!
According to tested build configurations, please try the following combinations
conda install -c conda-forge cudatoolkit=11.2 cudnn=8.1.0
Thank you!
@chunduriv That is exactly what I did, also I know for a fact I am using those versions, because I checked in the anaconda navigator.
I pasted the versions Iām using in my original post above.
Is there anything else I should look at?
The video i had linked recommended installing the CudaToolkit and cuDNN DLLs from NVIDIA website. Should i uninstall them?
1 Like
@hashbang,
Yes, i suspect issue may be due to difference in the minor versions.
Could you please share output of nvidia-smi ?
Thank you!
Here is the output:
>nvidia-smi
Fri Jun 16 17:02:06 2023
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 516.94 Driver Version: 516.94 CUDA Version: 11.7 |
|-------------------------------+----------------------+----------------------+
| GPU Name TCC/WDDM | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|===============================+======================+======================|
| 0 NVIDIA GeForce ... WDDM | 00000000:01:00.0 On | N/A |
| 0% 39C P8 16W / 198W | 711MiB / 8192MiB | 1% Default |
| | | N/A |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=============================================================================|
| 0 N/A N/A 7448 C+G C:\Windows\explorer.exe N/A |
| 0 N/A N/A 7900 C+G ...artMenuExperienceHost.exe N/A |
| 0 N/A N/A 8404 C+G ...2txyewy\TextInputHost.exe N/A |
| 0 N/A N/A 9416 C+G ...5n1h2txyewy\SearchApp.exe N/A |
| 0 N/A N/A 10792 C+G ...perience\NVIDIA Share.exe N/A |
| 0 N/A N/A 10816 C+G ...e\PhoneExperienceHost.exe N/A |
| 0 N/A N/A 11748 C+G ...in7x64\steamwebhelper.exe N/A |
| 0 N/A N/A 12100 C+G ...ge\Application\msedge.exe N/A |
| 0 N/A N/A 12316 C+G ...8wekyb3d8bbwe\Cortana.exe N/A |
| 0 N/A N/A 13504 C+G ...me\Application\chrome.exe N/A |
| 0 N/A N/A 14824 C+G ...\app-1.0.9013\Discord.exe N/A |
| 0 N/A N/A 15396 C+G ...icrosoft VS Code\Code.exe N/A |
| 0 N/A N/A 16148 C+G ...lPanel\SystemSettings.exe N/A |
+-----------------------------------------------------------------------------+
seems like im using CUDA 11.7, but when i try go to add/remove program i only see CUDA 11.2 related dependencies
Actually, i was able to install pytorch, since pytorch supports cuda 11.7.
I guess i can proceed with pytorch instead. Thanks!
1 Like