我需要与'multiprocessing'库并行运行多个类函数,但我无法找到针对我的具体问题的指南或答案.我的问题是只有第一个进程正在启动(或运行我不知道这里的确切区别).
为了说明这个问题,我设置了以下示例:
import multiprocessing
import time
class parallel_printer(multiprocessing.Process):
def __init__(self, freq_0, freq_1, freq_2):
self.freq_0 = freq_0
self.freq_1 = freq_1
self.freq_2 = freq_2
def print_0(self):
while True:
now = time.localtime()
if now.tm_sec % self.freq_0 == 0:
print('printer 0')
time.sleep(1.0)
def print_1(self):
while True:
now = time.localtime()
if now.tm_sec % self.freq_1 == 0:
print('printer 1')
time.sleep(1.0)
def print_2(self):
while True:
now = time.localtime()
if now.tm_sec % self.freq_2 == 0:
print('printer 2')
time.sleep(1.0)
def start_printer(self):
p_0 = multiprocessing.Process(target = self.print_0())
p_1 = multiprocessing.Process(target …Run Code Online (Sandbox Code Playgroud) 假设我有以下几个变量函数的简单示例
@tf.function
def f(A, Y, X):
AX = tf.matmul(A, X)
norm = tf.norm(Y - AX)
return norm
N = 2
A = tf.Variable(np.array([[1, 2], [3, 4]]))
Y = tf.Variable(np.identity(N))
X = tf.Variable(np.zeros((N, N)))
Run Code Online (Sandbox Code Playgroud)
我如何找到TensorflowX最小化?f我对一种通用解决方案感兴趣,该解决方案可与上面声明的函数一起使用,并且当有多个变量需要优化时。