یادگیری عمیق

دانشگاه فردوسی مشهد

محمود امین‌طوسی

Deep Learning

Mahmood Amintoosi

m.amintoosi @ um.ac.ir

پاییز ۱۴۰۲

Appendix-C

TensorFlow-GPU

NVIDIA GTX TITAN Black

TensorFlow-GPU

Installation

My Experience about OS replacement:

I had a working Anaconda TensorFlow-GPU environment on my previous Win10 OS. After Installing the new Win10 OS, NVIDIA Driver and Anaconda, I have had a working TensorFlow-GPU envoronment, by moving the previous TensorFlow-GPU environment to the new Anaconda :)

TensorFlow-GPU

Installation Check

NVIDIA Control Panle
C:\Program Files\NVIDIA Corporation\NVSMI\nvidia-smi.exe
CUDA-Z
nvidia-smi -L
conda list cudnn
numba -s

Installation Check: Python, Method 1

	 
		import tensorflow as tf
		with tf.Session() as sess:
			devices = sess.list_devices()
			print(devices)

Output in my device:
[_DeviceAttributes(/job:localhost/replica:0/task:0/device:CPU:0, CPU, 268435456), _DeviceAttributes(/job:localhost/replica:0/task:0/device:GPU:0, GPU, 3163186790)]

Installation Check: Python, Method 2

	 
		import tensorflow as tf
		if tf.test.gpu_device_name():
			print('Default GPU Device: {}'.format(tf.test.gpu_device_name()))
		else:
			print("Please install GPU version of TF")

Output in my device:
Default GPU Device: /device:GPU:0

Installation Check: Python, Method 3

	 
		from tensorflow.python.client import device_lib
		print(device_lib.list_local_devices())

Output in my device:
[name: "/device:CPU:0" device_type: "CPU" memory_limit: 268435456 locality { } incarnation: 17246399029107811532
, name: "/device:GPU:0" device_type: "GPU" memory_limit: 3163186790 locality { bus_id: 1 links { } } incarnation: 4513353947107249193 physical_device_desc: "device: 0, name: GeForce GTX 1050, pci bus id: 0000:01:00.0, compute capability: 6.1" ]

- Questions? -

m.amintoosi @ gmail.com

webpage : http://mamintoosi.ir

webpage in github : http://mamintoosi.github.io

github : mamintoosi