分析:已经排好序了,用二分查找最为合适
def search_range(arr, low, high, key):
out = [-1, -1]
mid = (low+high) // 2
if low == high & key != arr[low]:
return [-1, -1]
if key == arr[mid]:
for i in range(mid, high): # range()左闭右开
if key == arr[i]:
out[1] = i
j = mid
while arr[j] == key:
out[0] = j
j = j - 1
return out
elif key < arr[mid]:
temp_high = mid
return search_range(arr, low=low, high=temp_high, key=key)
else:
temp_low = mid + 1
return search_range(arr, low=temp_low, high=high, key=key)
if __name__ == '__main__':
list1 = [5, 7, 7, 8, 8, 10, 10]
print(search_range(list1, low=0, high=len(list1), key=10))
给定一个排序的数组和一个目标值,如果找到了目标,则返回索引。如果没有,则返回索引 如果按顺序插入,则返回其所在的索引。你可以假设数组中没有重复的东西。
方法一:循环遍历,这里不再赘述。
方法二:先排序,然后返回插入的值的 index
def searchInsert(nums, target):
if target not in nums:
nums.append(target) # 末尾追加
nums.sort() # 排序
return nums.index(target) # 返回索引
方法三:二叉搜索(推荐)
def search_insert(nums, low, high, key):
if low == high:
return low
mid = (high + low) // 2
if nums[mid] < key:
temp_low = mid + 1
return search_insert(nums, low=temp_low, high=high, key=key)
elif nums[mid] > key:
temp_high = mid
return search_insert(nums, low=low, high=temp_high, key=key)
else: # nums[mid] == key
return mid
if __name__ == '__main__':
list1 = [1, 3, 5, 7, 8, 10]
print(search_insert(list1, low=0, high=len(list1), key=9))
写一个有效的算法,在一个m×n的矩阵中搜索一个值。这个矩阵有以下特性。
例如,以下矩阵。
[[1,3,5,7],
[10,11,16,20],
[23,30,34,50]]
解法一:使用numpy将二维数组变为一维数组,然后使用二叉搜索
def search_2dm(matrix_flatten, low, high, target):
if low >= high and matrix_flatten[low] != target: # 没有key值
return False
mid = (high + low) // 2
if matrix_flatten[mid] > target:
temp_high = mid
return search_2dm(matrix_flatten, low=low, high=temp_high, target=target)
elif matrix_flatten[mid] < target:
temp_low = mid + 1
return search_2dm(matrix_flatten, low=temp_low, high=high, target=target)
else: # matrix_flatten[mid] == key
return True
if __name__ == '__main__':
list1 = [[1, 3, 5, 7],
[10, 11, 16, 20],
[23, 30, 34, 50]]
list1 = np.array(list1).flatten()
print(search_2dm(list1, low=0, high=len(list1), target=2))
方法二:
def search_matrix(matrix, target):
matrix = np.array(matrix)
row = matrix.shape[0] # 获取行号
column = matrix.shape[1] # 获取列数
first = 0
last = row * column
while first < last:
mid = first + (last-first)//2
value = matrix[(mid // column), (mid % column)]
if value == target:
return True
elif value < target:
first = mid + 1
else:
last = mid
return False
if __name__ == '__main__':
list1 = [[1, 3, 5, 7],
[10, 11, 16, 20],
[23, 30, 34, 50]]
print(search_matrix(list1, 70))