IndexError：索引超出范围-Java 学习之路

我是Python新手，感谢任何建设性的反馈 . 我的任务是使用ODEINT解决PDE，我的空间网格定义如下：

L = [8.0, 4.0, 8.0]   # Lenght of spatial zones 
dN = 1.0e2            # Number of grid points
N = [int(l*dN) for l in L] # Gives number of grid points over each spatial zone

通过使用ODEINT函数，我必须为我的解决方案定义一个零数组，以及一组时间点 . 我正在努力找出如何为数组写出正确的大小 . 我试过了：

dydt = np.zeros(shape=N) # For solution
t = np.linspace(0, 2, 2000) # for time

但是这给了我错误信息：

IndexError Traceback（最近一次调用最后一次）在（）70 t = np.linspace（0,2,1N）71 ---> 72 U = odeint（数字，y0，t，args =（h，D1，D2，N）））73 74 Fexit = U / Np

/Users/severinlangeberg/anaconda/lib/python3.5/site-packages/scipy/integrate/odepack.py in odeint（func，y0，t，args，Dfun，col_deriv，full_output，ml，mu，rtol，atol，tcrit ，h0，hmax，hmin，ixpr，mxstep，mxhnil，mxordn，mxords，printmessg）213 output = _odepack.odeint（func，y0，t，args，Dfun，col_deriv，ml，mu，214 full_output，rtol，atol，tcrit ，h0，hmax，hmin， - > 215 ixpr，mxstep，mxhnil，mxordn，mxords）216如果输出[-1] <0：217 warning_msg = _msgs [output [-1]]“运行full_output = 1得到量化信息 . “

在数字（y，t，h，D1，D2，N）39 40中对于[N [0] 1]中的i：＃800 =特殊2，必须从N [1] ---> 41 dydt的开始[ i] = D2 *（-3 * y [i] 4 * y [i 1] - y [i 2] / h ** 2）＃pluss 42 43 for i in range（N [0] 2，N [0 ] N [1] - 2）：＃从801到1197，或2到397

IndexError：索引801超出了轴0的大小为800的范围

Which seems to come from my loops:
# ODE setup
def numeric(y, t, h, D1, D2, N):

    dydt = np.zeros((N))

    # End grid points
    dydt[0] = 0
    dydt[-1] = 0


    # Inner grid points
    for i in range (1, N[0] - 2): # 1 to 797
        dydt[i] = D1 * (y[i - 1] - 2*y[i]+ y[i + 1]) / h**2 # range i = j - 1

    for i in [N[0] - 2]: # 798 = special 1
        dydt[i] =  D1 * (-3*y[i] - 4*y[i - 1] + y[i - 2]) / h**2 # minus

    for i in [N[0] - 1]: # 799 = unknown, and last index in N[0]
        dydt[i] = ((D1*(-3*y[i] + 4*y[i + 1] - y[i + 2])) - (D2*(-3*y[i] - 4*y[i - 1] + y[i - 2])))

    for i in [N[0] + 1]: # 800 = special 2, must be from begining of N[1]
        dydt[i] = D2 * (-3*y[i] + 4*y[i + 1] - y[i + 2] / h**2) # pluss

    for i in range (N[0] + 2, N[0] + N[1] - 2): # from 801 to 1197, or 2 to 397
        dydt[i] = D2 * (y[i - 1] - 2*y[i] + y[i + 1]) / h**2 # range

    for i in [N[0] + N[1] - 1]: # at 1198, or 398 = special 3
        dydt[i] = D2 * (-3*y[i] - 4*y[i - 1] + y[i - 2]) / h**2 # range

    for i in [N[0] + N[1]]: # 1199, or 399 = unknown
        dydt[i] = ((D1*(-3*y[i] + 4*y[i + 1] - y[i + 2])) - (D2*(-3*y[i] - 4*y[i - 1] + y[i - 2])))

    for i in [N[0] + N[1] + 1]: # at 1200, or 1 = special 4
        dydt[i] =  D1 * (-3*y[i + 1] + 4*y[i + 2] - y[i + 3] / h**2)  # pluss

    for i in range (N[0] + N[1] + 2, N[0] + N[1] + N[2] - 1): # 1202 to 1999
        dydt[i] = D1 * (y[i - 1] - 2*y[i]+ y[i + 1]) / h**2 # range

    return dydt

2 回答

从快速看，似乎问题是for循环，而不是使用 for i in [N[0]] ，你确定它不是 for i in range(N[0]) ？

for i in [N[0]] 相当多余，因为列表只包含一个元素，即 . N[0] ，所以当你试图寻找 dydt[800] <时，大小 800 的列表变得超出索引，因为最大索引是 799 .

像这样：

In [9]: N
Out[9]: [800, 400, 800]

In [10]: for i in [N[0]]:  # ie. i == N[0] == 800
   ....:     dydt[i] = 0
   ....:     
---------------------------------------------------------------------------
IndexError                                Traceback (most recent call last)
<ipython-input-10-04daa45ec403> in <module>()
      1 for i in [N[0]]:  # ie. i == N[0] == 800
----> 2     dydt[i] = 0
      3 

IndexError: index 800 is out of bounds for axis 0 with size 800

In [11]: for i in range(N[0]):
   ....:     dydt[i] = 0
   ....:

回复于 2024-04-27T16:58:12+08:00

L = [8.0, 4.0, 8.0] 
dN = 1.0e2 
N = [int(l*dN) for l in L] # create array [800, 400, 800] 

dydt = np.zeros(shape=N) # create 3 dimensional matrix 
                         # 1d has size 800 (from 0 to 799 including)
                         # 2d has size 400 (from 0 to 399 including)
                         # 3d same as 1st

在这段代码中：

for i in[N[0]]: 
    dydt[i] = 0 # Unknown point

我总是800，所以它实际上与 dydt[ 800 ] = 0 相同 . 这比799大 .

看起来你用错误的方式使用numpy数组 .

是的，在低级别它们代表连续的一系列整数或浮动的任何东西 . 但是在python级别上，np.array表示为数组数组 . （即第一个元素为3d矩阵 mtx = np.zeros(shape=[800,400,800] 在python中将是 mtx[0][0][0] ）） .

所以 for i in[N[0]]: 之后的所有代码都不起作用 . 它试图访问超出矩阵范围的子数组 .

更新：

例如，你有3d矩阵（3x3x3）

mtx=np.zero(shape=[3,3,3])
 print(mtx) 
 #[ [ [ 0.  0.  0.]
 #    [ 0.  0.  0.]
 #    [ 0.  0.  0.] ]

 #  [ [ 0.  0.  0.]
 #    [ 0.  0.  0.]
 #    [ 0.  0.  0.] ]

 #  [ [ 0.  0.  0.]
 #    [ 0.  0.  0.]
 #    [ 0.  0.  0.] ] ]

 mtx[0] = 1 
 print( mtx )
 #[ [ [ 1.  1.  1.]
 #    [ 1.  1.  1.]
 #    [ 1.  1.  1.] ]

 #  [ [ 0.  0.  0.]
 #    [ 0.  0.  0.]
 #    [ 0.  0.  0.] ]

 #  [ [ 0.  0.  0.]
 #    [ 0.  0.  0.]
 #    [ 0.  0.  0.] ] ]

 mtx[0][1] = 2
 #[ [ [ 1.  1.  1.]
 #    [ 2.  2.  2.]
 #    [ 1.  1.  1.] ]

 # others are zeros

 mtx[0][1][2] = 3
 #[ [ [ 1.  1.  1.]
 #    [ 2.  2.  3.]
 #    [ 1.  1.  1.] ]

 # others are zeros

回复于 2024-04-27T16:58:12+08:00

IndexError：索引超出范围

但是这给了我错误信息：

2 回答

相关问题