Python 如何解決稀疏矩陣運算

用Python求解微分線性方程

因為之前用matlab也編寫過,所以前不久試著用python寫,感覺之間互通點也蠻多的,易理解。

題目:稀疏線性方程組的求解方法

簡單的方程如: AX=b

其中

A矩陣與b向量

python有很多功能庫,這些庫對於編程很有幫助,可以在pycharm的Project Interpreter導入庫,例如numpy、os、scipy等比較基礎的庫,

下面是用來求解的代碼:

import numpy as np
from scipy import linalg
import os
#輸入矩陣維數
print("你好,這裡是計算稀疏矩陣線性方程組的地方,非誠勿擾!")
dism_num = input("你的A矩陣維數是:")
dism_num = int(dism_num)
print("接下來請你依次輸入矩陣的行向量(註意隻能輸入英文逗號,):")
A =[]
#X =[]
for i in range(1,dism_num+1):
    a=input("第"+str(i)+"行向量是:")
    alist = a.split(",")
    alist = [int(alist[j]) for j in range(len(alist))]
    A.append(alist)
print("你所輸入的矩陣行向量是:")
print(A)
#記錄輸入的X矩陣

#輸入向量b
print("輸入b向量")
b = input("b向量是:")
b_list = b.split(",")
b_list = [int(b_list[j]) for j in range(len(b_list))]
print("你輸入的b向量是:")
print(b_list)
#記錄b向量

#詢問是否計算單個答案(單元素)
ask = input("是否隻需求解單個值:(是或否)")
while(True):
    if ask == '是':
        ask_a = 'T'
        ask_num = input("請繼續輸入你所需要的答案序號:")
        ask_num = int(ask_num)
        if ask_num<=dism_num and ask_num>0:
            print("OK,馬上幫你計算")
            break
        else:
            print("輸入的值超出矩陣維數,請重新輸入:")
    if ask == '否':
        ask_a = 'F'
        break
#詢問完成,隻有當用戶輸入正確的序號才可以進行計算,否則重新詢問

#開始計算x向量瞭
A = np.array(A)
b = np.array(b_list)
x = linalg.solve(A,b)
print("計算的結果的:")
if ask_a == 'F':
    print(x)
if ask_a =='T':
    print(x[ask_num-1])
#計算完x向量瞭

os.system("pause")
#用於py文件結束玩暫停顯示結果

其基本流程如圖:

代碼開發流程

運行結果如下:

補充:python 多線程稀疏矩陣乘法

看代碼吧~

import threading, time
import numpy as np
res = []
class MyThread(threading.Thread):
    def __init__(self,i,j,m1,m2):
        threading.Thread.__init__(self)
        self.x, self.y = i,j
        self.m1, self.m2 = m1, m2
    def run(self):
        global res, lock
        if lock.acquire():
            m1 = self.m1[self.m1[:,0]==self.x]
            m2 = self.m2[self.m2[:,1]==self.y]
            value = 0.
            for item1 in m1:
                for item2 in m2:
                    if item1[1] == item2[0]:
                        value += item1[2]*item2[2]
            res.append([self.x,self.y,value])
            lock.release()
if "__main__" == __name__:
    m1 = [[2,2],[0,0,1],[0,1,2],[1,0,3],[1,1,4]]
    m2 = [[2,3],[0,0,2],[0,2,1],[1,2,3],[1,1,4]]
    s1, s2 = m1[0], m2[0]
    assert s1[1]==s2[0], 'mismatch'
    m1_value = np.array(m1[1:])
    m2_value = np.array(m2[1:])
    rows, cols = s1[0], s2[1]
    res.append([rows, cols])
    ThreadList = []
    lock = threading.Lock()
    for i in range(rows):
        for j in range(cols):
            t = MyThread(i,j,m1_value,m2_value)
            ThreadList.append(t)
    for t in ThreadList:
        t.start()
    for t in ThreadList:
        t.join()
    print (res)

以上為個人經驗,希望能給大傢一個參考,也希望大傢多多支持WalkonNet。

推薦閱讀: