在 n 级列中逐行分类 excel 数据

Question

我在使用 excel 文件对某些列和行中的数据进行分类时遇到问题，我需要将合并单元格安排到下一列作为 1 行，然后下一列像下面的图片一样放在它们旁边：

输入：

乳制品产量：

总结：

首先我们取Dairy行，然后我们转到Dairy前面的第二列，得到Dairy前面的数据，然后我们转到第二列，然后在 Milk to Mr. 1 前面我们得到 Butter to Mrs. 1 和 Butter to Mrs. 2 等等...

之后我们想将它导出到一个 excel 文件中，就像在输出图片中一样。

我写了一段代码，它获取第一列数据并找到它前面的所有数据，但我需要更改它以便像输出图片中那样逐行获取数据：

import pandas
import openpyxl
import xlwt
from xlwt import Workbook

df = pandas.read_excel('excel.xlsx')

result_first_level = []

for i, item in enumerate(df[df.columns[0]].values, 2):
    if pandas.isna(item):
        result_first_level[-1]['index'] = i
    else:
        result_first_level.append(dict(name=item, index=i, levels_name=[]))

for level in df.columns[1:]:
    move_index = 0
    for i, obj in enumerate(result_first_level):
        if i == 0:
            for item in df[level].values[0:obj['index'] - 1]:
                if pandas.isna(item):
                    move_index += 1
                    continue
                else:
                    obj['levels_name'].append(item)
                move_index += 1
        else:
            for item in df[level].values[move_index:obj['index'] - 1]:
                if pandas.isna(item):
                    move_index += 1
                    continue
                else:
                    obj['levels_name'].append(item)
                move_index += 1

# Workbook is created
wb = Workbook()

# add_sheet is used to create sheet.
sheet1 = wb.add_sheet('Sheet 1')
style = xlwt.easyxf('font: bold 1')

move_index = 0
for item in result_first_level:
    for member in item['levels_name']:
        sheet1.write(move_index, 0, item['name'], style)
        sheet1.write(move_index, 1, member)
        move_index += 1

wb.save('test.xls')

下载Input File excel from here

感谢您的帮助！

Answer 1

首先，填写您的数据以使用最后一个有效值填充空白单元格，然后使用 pd.CategoricalDtype 创建一个有序集合以对 product 列进行排序。最后，您只需成对迭代列并重命名列以允许连接。最后一步是按 product 值对行进行排序。

import pandas as pd

# Prepare your dataframe
df = pd.read_excel('input.xlsx').dropna(how='all')
df.update(df.iloc[:, :-1].ffill())
df = df.drop_duplicates()

# Get keys to sort data in the final output
cats = pd.CategoricalDtype(df.T.melt()['value'].dropna().unique(), ordered=True)

# Group pairwise values
data = []
for cols in zip(df.columns, df.columns[1:]):
    col_mapping = dict(zip(cols, ['product', 'subproduct']))
    data.append(df[list(cols)].rename(columns=col_mapping))

# Merge all data
out = pd.concat(data).drop_duplicates().dropna() \
        .astype(cats).sort_values('product').reset_index(drop=True)

输出：

>>> cats
CategoricalDtype(categories=['Dairy', 'Milk to Mr.1', 'Butter to Mrs.1',
                  'Butter to Mrs.2', 'Cheese to Miss 2 ', 'Cheese to Mr.2',
                  'Milk to Miss.1', 'Milk to Mr.5', 'yoghurt to Mr.3',
                  'Milk to Mr.6', 'Fruits', 'Apples to Mr.6',
                  'Limes to Miss 5', 'Oranges to Mr.7', 'Plumbs to Miss 5',
                  'apple for mr 2', 'Foods & Drinks', 'Chips to Mr1',
                  'Jam to Mr 2.', 'Coca to Mr 5', 'Cookies to Mr1.',
                  'Coca to Mr 7', 'Coca to Mr 6', 'Juice to Miss 1',
                  'Jam to Mr 3.', 'Ice cream to Miss 3.', 'Honey to Mr 5',
                  'Cake to Mrs. 2', 'Honey to Miss 2',
                  'Chewing gum to Miss 7.'], ordered=True)

>>> out
             product              subproduct
0              Dairy            Milk to Mr.1
1              Dairy          Cheese to Mr.2
2       Milk to Mr.1         Butter to Mrs.1
3       Milk to Mr.1         Butter to Mrs.2
4    Butter to Mrs.2       Cheese to Miss 2 
5     Cheese to Mr.2          Milk to Miss.1
6     Cheese to Mr.2         yoghurt to Mr.3
7     Milk to Miss.1            Milk to Mr.5
8    yoghurt to Mr.3            Milk to Mr.6
9             Fruits          Apples to Mr.6
10            Fruits         Oranges to Mr.7
11    Apples to Mr.6         Limes to Miss 5
12   Oranges to Mr.7        Plumbs to Miss 5
13  Plumbs to Miss 5          apple for mr 2
14    Foods & Drinks            Chips to Mr1
15    Foods & Drinks         Juice to Miss 1
16    Foods & Drinks          Cake to Mrs. 2
17      Chips to Mr1            Jam to Mr 2.
18      Chips to Mr1         Cookies to Mr1.
19      Jam to Mr 2.            Coca to Mr 5
20   Cookies to Mr1.            Coca to Mr 6
21   Cookies to Mr1.            Coca to Mr 7
22   Juice to Miss 1           Honey to Mr 5
23   Juice to Miss 1            Jam to Mr 3.
24      Jam to Mr 3.    Ice cream to Miss 3.
25    Cake to Mrs. 2  Chewing gum to Miss 7.
26    Cake to Mrs. 2         Honey to Miss 2

在 n 级列中逐行分类 excel 数据

classifing excel data row by row in n level columns

python

excel

dataset

python-3.x

pandas