使用 python 在二维列表中搜索以查找 x,y 位置

Question

I have 2D list and I need to search for the index of an element. As I am begineer to programming I used the following function:

def in_list(c):
    for i in xrange(0,no_classes):
        if c in classes[i]:
            return i;

    return -1

Here classes is a 2D list and no_classes denotes the number of classes ie the 1st dimesntion of the list. -1 is returned when c is not in the araray. Is there any I can optimize the search?

Answer 1

You don't need to define no_classes yourself. Use enumerate() :

def in_list(c, classes):
    for i, sublist in enumerate(classes):
        if c in sublist:
            return i
    return -1

Answer 2

Use list.index(item)

a = [[1,2],[3,4,5]]

def in_list(item,L):
    for i in L:
        if item in i:
            return L.index(i)
    return -1

print in_list(3,a)
# prints 1

Answer 3

if order doesn't matter and you have no duplicates in your data, I suggest to turn you 2D list into list of sets:

>>> l = [[1, 2, 4], [6, 7, 8], [9, 5, 10]]
>>> l = [set(x) for x in l]
>>> l
[set([1, 2, 4]), set([8, 6, 7]), set([9, 10, 5])]

After that, your original function will work faster, because search of element in set is constant (while search of element in list is linear), so you algorithm becomes O(N) and not O(N^2).

Note that you should not do this in your function or it would be converted each time function is called.

Answer 4

If your "2D" list is rectangular (same number of columns for each line), you should convert it to a numpy.ndarray and use numpy functionalities to do the search. For an array of integers, you can use == for comparison. For an array of float numbers, you should use np.isclose instead:

a = np.array(c, dtype=int)
i,j = np.where(a == element)

or

a = np.array(c, dtype=float)
i,j = np.where(np.isclose(a, element))

such that i and j contain the line and column indices, respectively.

Example:

a = np.array([[1, 2],
              [3, 4],
              [2, 6]], dtype=float)
i, j = np.where(np.isclose(a, 2))
print(i)
#array([0, 2])
print(j)
#array([1, 0]))