根据相同的标题前缀从多个 DataFrame 中选择列

Question

I have a function that iterates over the rows of a csv for the Age column and if an age is negative, it will print the Key and the Age value to a text file.

def neg_check():
    results = []

    file_path = input('Enter file path: ')
    file_data = pd.read_csv(file_path, encoding = 'utf-8')
    
    for index, row in file_data.iterrows():
        if row['Age'] < 0:
            results.append((row['Key'], row['Age']))
    with open('results.txt', 'w') as outfile:
        outfile.write("\n".join(map(str, results)))   
        outfile.close()

In order to make this code repeatable, how can I modify it so it will iterate the rows if the column starts with " Age "? My files have many columns that start with " Age " but end differently. . I tried the following...

if row.startswith['Age'] < 0:

and

if row[row.startswith('Age')] < 0:

but it throws AttributeError: 'Series' object has no attribute 'startswith' error.

My csv files:

sample 1

Key   Sex     Age
    1        Male          46
    2        Female        34

sample 2

Key   Sex     AgeLast
    1        Male          46
    2        Female        34

sample 3

Key   Sex     AgeFirst
    1        Male          46
    2        Female        34

Answer 1

I would do this in one step, but there are a few options. One is filter :

v = df[df.filter(like='AgeAt').iloc[:, 0] < 0]

Or,

c = df.columns[df.columns.str.startswith('AgeAt')][0]
v = df[df[c] < 0]

Finally, to write to CSV, use

if not v.empty:
    v.to_csv('invalid.csv')

Looping over your data is not necessary with pandas.