Скачать или смотреть Install Tensorflow with GPU support on windows

To install TensorFlow with CUDA support, follow these step-by-step instructions.

TensorFlow 2.10 was the last TensorFlow release that supported GPU on native-Windows.
Ref: https://www.tensorflow.org/install/pi...


Download and install cuda and cudnn - Match TensorFlow version with the appropriate CUDA and cuDNN versions
https://www.tensorflow.org/install/so...

We are downloading tensorflow 2.10 which required cuda 11.2 and cudnn 8.1

pip install tensorflow==2.10.0


How to Confirm GPU Usage?

import tensorflow as tf

Check if TensorFlow is using the GPU
if tf.config.list_physical_devices('GPU'):
print("TensorFlow is using the GPU!")

Get list of available GPUs
gpus = tf.config.list_physical_devices('GPU')
print("Available GPUs:", gpus)

Get details of each GPU
for gpu in gpus:
details = tf.config.experimental.get_device_details(gpu)
print("GPU Details:", details)
else:
print("TensorFlow is using the CPU.")


Output:
TensorFlow is using the GPU!
Available GPUs: [PhysicalDevice(name='/physical_device:GPU:0', device_type='GPU')]
GPU Details: {'device_name': 'NVIDIA GeForce RTX 3090', 'compute_capability': (8, 6)}


#tensorflow #computervision #cuda #nvidia