如何在导出到 Excel 文件之前从 Python DataFrame 设置多行样式

Posted 2023-03-11

【中文标题】如何在导出到 Excel 文件之前从 Python DataFrame 设置多行样式

【英文标题】：How to style several rows from Python DataFrame before exporting to Excel file

【发布时间】：2021-06-10 16:32:22

【问题描述】：

我希望我的 DataFrame 最后 3 行打印为粗体。下面是我的 DataFrame 结构，包括文本和图像（从 Jupyter Notebook 输出）格式。

# Actually, there are some more columns but it's not necessary to put them all here. 
# These columns are enough to give an example.
      epoch      loss  accuracy      f1_m  precision_m  recall_m     lr  
1       1.0  5.104000  0.494600  0.458400     0.588500  0.403900  0.001   
2       2.0  3.313200  0.801000  0.799500     0.828500  0.772600  0.001   
3       3.0  3.063600  0.825000  0.824400     0.852900  0.797700  0.001   
4       4.0  2.855900  0.838900  0.838600     0.865700  0.813100  0.001   
5       5.0  2.659400  0.847500  0.846100     0.876500  0.817800  0.001   
6       6.0  2.490200  0.851100  0.851600     0.876800  0.827900  0.001   
7       7.0  2.334100  0.854100  0.853400     0.880100  0.828200  0.001   
8       8.0  2.168300  0.862000  0.861800     0.885700  0.839100  0.001   
9       9.0  2.025500  0.864200  0.863800     0.888800  0.840300  0.001   
10     10.0  1.900300  0.864100  0.863900     0.889100  0.840200  0.001   
11     11.0  1.774600  0.866300  0.866200     0.888800  0.844700  0.001   
12     12.0  1.668100  0.867000  0.866000     0.891700  0.841700  0.001   
13     13.0  1.557800  0.869900  0.869600     0.893500  0.846900  0.001   
14     14.0  1.459500  0.871900  0.872000     0.895900  0.849400  0.001   
15     15.0  1.386500  0.869000  0.868600     0.892800  0.845700  0.001   
16     16.0  1.294400  0.872900  0.873000     0.895800  0.851400  0.001   
17     17.0  1.210600  0.876000  0.875700     0.898500  0.853900  0.001   
18     18.0  1.156100  0.875800  0.875600     0.897800  0.854500  0.001   
19     19.0  1.100800  0.872800  0.872100     0.897200  0.848300  0.001   
20     20.0  1.042000  0.875000  0.875200     0.897100  0.854400  0.001   
21     21.0  0.991800  0.875700  0.876400     0.899100  0.854900  0.001   
22     22.0  0.942500  0.878600  0.878300     0.900800  0.856900  0.001   
23     23.0  0.896600  0.878800  0.878400     0.901900  0.856200  0.001   
24     24.0  0.857100  0.880500  0.880200     0.901800  0.859700  0.001   
25     25.0  0.826900  0.878600  0.878400     0.900600  0.857300  0.001   
26     26.0  0.795000  0.878600  0.878100     0.901900  0.855600  0.001   
27     27.0  0.778700  0.876300  0.875800     0.900400  0.852500  0.001   
28     28.0  0.750800  0.876100  0.875700     0.899300  0.853300  0.001   
29     29.0  0.718600  0.878900  0.878900     0.901900  0.857000  0.001   
30     30.0  0.707000  0.876000  0.875400     0.899100  0.852900  0.001   
31     31.0  0.694900  0.875800  0.875900     0.900000  0.853100  0.001   
32     32.0  0.665100  0.879300  0.879100     0.900600  0.858600  0.001   
33     33.0  0.657700  0.875600  0.875400     0.899700  0.852500  0.001   
34     34.0  0.637700  0.878100  0.878100     0.901000  0.856300  0.001   
35     35.0  0.618600  0.880100  0.879500     0.902900  0.857300  0.001   
36     36.0  0.601200  0.880400  0.880600     0.902700  0.859600  0.001   
37     37.0  0.589900  0.882300  0.882100     0.903400  0.861800  0.001   
38     38.0  0.580500  0.880400  0.880600     0.901800  0.860400  0.001   
39     39.0  0.573000  0.880200  0.880300     0.902100  0.859600  0.001   
40     40.0  0.555900  0.882200  0.881700     0.904700  0.859900  0.001   
41     41.0  0.551200  0.880700  0.880300     0.902900  0.858800  0.001   
42     42.0  0.547500  0.880400  0.879900     0.903300  0.857600  0.001   
43     43.0  0.531100  0.883500  0.883500     0.905500  0.862600  0.001   
44     44.0  0.537000  0.880000  0.879500     0.903300  0.856900  0.001   
45     45.0  0.526200  0.883200  0.882200     0.904300  0.861200  0.001   
46     46.0  0.517900  0.882000  0.882300     0.903000  0.862500  0.001   
47     47.0  0.514400  0.881100  0.880900     0.902100  0.860700  0.001   
48     48.0  0.505000  0.883400  0.882700     0.904800  0.861600  0.001   
49     49.0  0.499200  0.883300  0.883600     0.904600  0.863500  0.001   
50     50.0  0.491200  0.884600  0.884700     0.905600  0.864800  0.001   
min     1.0  0.491200  0.494600  0.458400     0.588500  0.403900  0.001   
max    50.0  5.104000  0.884600  0.884700     0.905600  0.864800  0.001   
mean   25.5  1.285006  0.857558  0.855906     0.883479  0.832423  0.001  

对于初学者，我尝试使用来自this 邮政编码的 2 个答案来测试它是否适用于最后一行，但它没有用。

还有其他想法来设置 DataFrame 的样式吗？

以防万一，我在这里附上我当前的完整代码：

import fileinput
import glob
import pathlib
import pandas as pd
import os
import numpy as np

def readText(filename):
    with open(filename, "r") as f:
        data = f.readlines()
    return data

# I want my DataFrame to be exported to Excel files later on
def toExcelFiles(filename, data_text):
    data = []
    for i, line in enumerate(data_text.split('\n')[1::2], start=1):
        obj = 
        obj['epoch'] = i
        for x in line.split(' - ')[2:]:
            k, v = x.split(':')
            obj[k] = v.lstrip()
        data.append(obj)

    df = pd.DataFrame(data).astype(float)
    filename = filename.replace(".txt", ".xlsx")
    print(filename)
    df.index = np.arange(1, len(df) + 1)
    df.loc['min'] = df.min()
    df.loc['max'] = df.max()
    df.loc['mean'] = df.mean()
    df.style.applymap('font-weight: bold', subset=pd.IndexSlice[df.index[df.index=='mean'], :])
    with pd.option_context('display.max_rows', None, 'display.max_columns', None):  # more options can be specified also
        print(df)
    df.to_excel(filename) 

# main
os.chdir(f'C:/Users/Dionisius/OneDrive/testresult/arsi 4/txt')
file_list = glob.glob("*.txt")
print("File_list: ", file_list)
print("*"*127)

for file in file_list:
    text = readText(file) # returns list
    string_epoch = ''.join(text) # mengubah list jadi string
    toExcelFiles(file, string_epoch)

.txt 的示例文件附在here 上。

【参考方案1】：

给你：

def bold_last(df, numrows=3):
    ret = pd.DataFrame('', index=df.index, columns=df.columns)

    # last few rows
    ret.iloc[-numrows:] = 'font-weight:bold'

    return ret

# pass numrows=4 to `apply` to bold 4 last rows
df.style.apply(bold_last, axis=None).to_excel('some_file.xlsx')

最后几行看起来像这样：

【讨论】：

.iloc[-:] 代码正是我正在寻找的。现在我还可以使用background-color:yellow 设置最后一行的样式。谢谢！