1、测试PyTorch能正常使用GPU:
import torch torch.cuda.is_available()
返回结果:
True
2、测试Tensorflow能正常使用GPU:
示例一:
tf.test.is_gpu_available()
返回结果:
True
示例二:
gpu_device_name = tf.test.gpu_device_name() print(gpu_device_name)
返回结果:
/device:GPU:0
示例三:
import tensorflow as tf with tf.device(‘/cpu:0‘): a = tf.constant([1,2,3],shape=[3],name=‘a‘) b = tf.constant([1,2,3],shape=[3],name=‘b‘) with tf.device(‘/gpu:0‘): c = a + b print(c)
返回结果:
tf.Tensor([2 4 6], shape=(3,), dtype=int32)
示例四:
import tensorflow.compat.v1 as tf tf.disable_v2_behavior() def performanceTest(device_name,size): with tf.device(device_name): #选择具体的哪一个设备 W = tf.random_normal([size, size],name=‘W‘) #以随机产生数值的方式建立W矩阵 X = tf.random_normal([size, size],name=‘X‘) #以随机产生数值的方式建立X矩阵 mul = tf.matmul(W, X,name=‘mul‘) sum_result = tf.reduce_sum(mul,name=‘sum‘) #将mul矩阵里面的值加总 startTime = time.time() #记录开始运行的时间 tfconfig=tf.ConfigProto(log_device_placement=True) #代表显示设备的相关信息 with tf.Session(config=tfconfig) as sess: result = sess.run(sum_result) takeTimes=time.time() - startTime print(device_name," size=",size,"Time:",takeTimes ) return takeTimes #返回运行时间 g=performanceTest("/gpu:0",10000) c=performanceTest("/cpu:0",10000)
返回结果:
Device mapping: /job:localhost/replica:0/task:0/device:XLA_CPU:0 -> device: XLA_CPU device /job:localhost/replica:0/task:0/device:GPU:0 -> device: 0, name: GeForce RTX 2070 SUPER, pci bus id: 0000:01:00.0, compute capability: 7.5 /job:localhost/replica:0/task:0/device:XLA_GPU:0 -> device: XLA_GPU device /gpu:0 size= 10000 Time: 0.3135712146759033 Device mapping: /job:localhost/replica:0/task:0/device:XLA_CPU:0 -> device: XLA_CPU device /job:localhost/replica:0/task:0/device:GPU:0 -> device: 0, name: GeForce RTX 2070 SUPER, pci bus id: 0000:01:00.0, compute capability: 7.5 /job:localhost/replica:0/task:0/device:XLA_GPU:0 -> device: XLA_GPU device /cpu:0 size= 10000 Time: 5.693259954452515
参考:
https://blog.csdn.net/weixin_41770169/article/details/91437100
https://blog.csdn.net/william_hehe/article/details/79615894
https://www.runoob.com/python/python-date-time.html
https://blog.csdn.net/weixin_41503009/article/details/86930657
https://clatterrr.com/archives/2362
https://blog.csdn.net/weixin_38410551/article/details/103631977
原文地址:https://www.cnblogs.com/ratels/p/12397381.html
时间: 2024-11-06 07:11:32