错误数据类型的 Python 比较运算符

Question

我试图允许用户从 DataFrame 和比较运算符（>=、<= 或 ==）输入列名，并根据用户选择的运算符，将 DataFrame 值与他们选择的收盘价。 df.loc[df[column_name]] 部分和 valueofclose 部分工作正常，但是我收到一个错误：

numpy.core._exceptions.UFuncTypeError: ufunc 'add' did not contain a loop with signature matching types (dtype('float64'), dtype('<U2')) -> None

我认为问题是因为用户输入了一个字符串，我需要它成为源代码，我想到的唯一方法是在它周围使用 + 并添加 exec 函数，但这也会执行代码我不知道之后该怎么办。

valueofclose = int(input("What is your value of the percent close: "))

column_name = input("Which column would you like to focus on: Percent Gain, Percent Loss or Day % Change: ")
operator = (input("What would you like for the operator to be: >=, <= or =="))

print(type(operator))

 df_loss = df.loc[df[column_name] + operator + valueofclose] #lets me find days where percent loss is greater than 4
for i, rows in df_loss.iterrows(): #Goes through the df_loss dataframe (where percent is greater than 4)
my_listloss = rows.notindex - 1 #and gets the index of that and subtracts the index by 1 to get the index of the next day
Percentloss_list.append(my_listloss) #adds that index to the list

Answer 1

我希望这个解决方案对您有所帮助：

def add_op(a, b):
    return a + b

def power_op(a, b):
    return pow(a, b)

operation_dict = {
    '+': add_op,
    'power': power_op,
}

a, op, b = input("Enter command: ").split()
print(operation_dict[op](int(a), int(b)))

Enter command: 3 + 3
6

Enter command: 3 power 3
27

或者，在更简单的代码中使用lambda：

operation_dict = {
    '+': lambda a, b: a + b,
    'power': lambda a, b: pow(a, b),
}

a, op, b = input("Enter command: ").split()
print(operation_dict[op](int(a), int(b)))