Followed GPU installation but keep getting CPU install

General Discussion
#1

I have created a conda environment & then followed pip install instructions for Windows for Tensorflow GPU.

To test the install,

>python
>>> import tensorflow as tf
>>> tf.__version
'2.11.0'
>>> print(tf.config.list_local_devices())
[]
>>> from tensorflow.python.client import device_lib
>>> print(device_lib.list_local_devices())
2023-01-18 09:02:25.549619: 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:  AVX AVX2
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
[name: "/device:CPU:0"
device_type: "CPU"
memory_limit: 268435456
locality {
}
incarnation: 11568165271890850111
xla_global_id: -1
]

System: Windows 10 - 10.0.19044, x64, NVIDIA GeForce GTX 1050, Python 3.10.6

Am I right that that is a CPU version of TF? How do I get a GPU version given that I followed the instructions to install a GPU version but keep getting a CPU version?

#3

@DrBwts,

TensorFlow 2.10 was the last TensorFlow release that supported GPU on native-Windows. Starting with TensorFlow 2.11 , you will need to install TensorFlow in WSL2.

Could you follow step by step instructions as mentioned here and let us know?

After TensorFlow instalation, we can verify the GPU setup:

import tensorflow as tf
physical_devices = tf.config.list_physical_devices('GPU')
print("Num GPUs:", len(physical_devices))

Thank you!