在Pandas DataFrame中调整Excel来源Numpy数组列顺序报错解决

问题描述 #

我通过Pandas将Excel数据导入为Numpy数组，需将列顺序从['A','B','C','D','E','F','G','H']调整为['C','B','A','D','F','H','E','G']，但运行代码时出现“DataFrame object is not callable”错误，相关代码如下：

# importing python libraries
import pandas as pd
import numpy as np

# reading source excel files and storing into two DataFrames
df1 = pd.read_excel('filepath')

# reading data by row
datax = []

def getRows(excel_file, sheet_name, start_row = 1):
    excel_data = pd.read_excel(excel_file, sheet_name = sheet_name, header=None)
    #print(excel_data)
    size = excel_data.shape
    #print(size)
    row_num = size[0]
    for i in range(start_row, row_num):
        data = excel_data.iloc[i].values.tolist()
        datax.append(data)
    return datax

getRows("filepath", "sheetname", 0)

a = np.array(datax)

df = pd.DataFrame(a[:,[0,1,2,4,6,9,13,14]], columns = ['A','B','C','D','E','F','G','H'])


df2 = df(pd.DataFrame(a[:,[0,1,2,4,6,9,13,14]], columns =['C','B','A','D','F','H','E','G']))
df2.to_excel("filepath")

print(df2)

错误原因 #

• 核心错误在df2 = df(pd.DataFrame(...))这行：df是已经创建好的DataFrame对象，你却用函数调用的语法（括号）去操作它，这就触发了"DataFrame object is not callable"错误。
• 代码存在冗余：已经将数组转为df，没必要重复基于原数组创建新DataFrame来调整列顺序。

修正方案 #

调整DataFrame列顺序的最简方式，是直接用目标列名列表对已有DataFrame进行索引，不需要重复创建DataFrame。

优化后的完整代码 #

import pandas as pd
import numpy as np

# 读取Excel数据并按行提取
datax = []

def getRows(excel_file, sheet_name, start_row=0):
    excel_data = pd.read_excel(excel_file, sheet_name=sheet_name, header=None)
    row_num = excel_data.shape[0]
    for i in range(start_row, row_num):
        data = excel_data.iloc[i].values.tolist()
        datax.append(data)
    return datax

getRows("filepath", "sheetname", 0)

a = np.array(datax)

# 创建初始DataFrame，指定列名
df = pd.DataFrame(a[:,[0,1,2,4,6,9,13,14]], columns=['A','B','C','D','E','F','G','H'])

# 直接通过列名列表调整顺序（核心修正）
df2 = df[['C','B','A','D','F','H','E','G']]

# 导出到Excel，建议加上index=False避免导出索引列
df2.to_excel("output_filepath.xlsx", index=False)
print(df2)

进一步简化代码 #

可以去掉冗余的Numpy数组转换和全局变量，直接用Pandas完成所有操作：

import pandas as pd

# 直接读取Excel，选择需要的列并指定初始列名
df = pd.read_excel("filepath", sheet_name="sheetname", header=None)
selected_df = df.iloc[:, [0,1,2,4,6,9,13,14]]
selected_df.columns = ['A','B','C','D','E','F','G','H']

# 调整列顺序
df2 = selected_df[['C','B','A','D','F','H','E','G']]

# 导出结果
df2.to_excel("output_filepath.xlsx", index=False)
print(df2)

内容的提问来源于stack exchange，提问作者MeiLin M