如何通过索引访问DataFrame多列查找Excel行字符串及解决find报错问题

Hey there! Let's break down your two pandas-related questions and solve them step by step:

If your goal is to access multiple columns via row index and convert the row's content to a string, or find rows where any of the selected columns contain a specific string, here are two common approaches:

Approach 1: Fetch specified columns of a target index and concatenate into a string #

Suppose you have a DataFrame df loaded from Excel, and you want to access the row at target_index, select specific columns (like ['Function Name', 'Description', 'Status']), then join the row's values into a single string:

import pandas as pd

# Load data from Excel
df = pd.read_excel("your_file.xlsx")

target_index = 3  # Example target row index
selected_columns = ['Function Name', 'Description', 'Status']

# Access via loc, then convert to string
row_str = ' '.join(str(val) for val in df.loc[target_index, selected_columns])
print(row_str)

If you want the entire row (all columns) as a string, simply omit the column list:

row_str = ' '.join(str(val) for val in df.loc[target_index])

Approach 2: Find rows where any selected column contains a specific string #

To filter rows where any of the specified columns has your target string, use str.contains with any(axis=1):

target_str = "PowerMode"
selected_columns = ['Function Name', 'Description']

# Filter rows where any selected column contains the target string
matching_rows = df[df[selected_columns].apply(lambda x: x.str.contains(target_str, na=False)).any(axis=1)]

The na=False parameter handles empty values to avoid errors.

This error happens because your df['Function Name'] column contains float values—most commonly NaN (empty cells, which pandas automatically treats as float-type NaN). The find() method only works on strings, so when your code hits a float cell, it throws an error.

Even though you mentioned the column has content like "Power is applied in the lower stage", there must be empty cells (NaN) mixed in the column, causing the type mismatch.

Here are two solutions:

Solution 1: Convert the column to string type and handle empty values #

Convert the entire column to strings, either turning NaN into the string "nan" or replacing NaN with empty strings (the latter is more recommended):

# Option A: Convert directly, NaN becomes "nan"
df['Function Name'] = df['Function Name'].astype(str)

# Option B: Replace NaN with empty strings first (better for consistency)
df['Function Name'] = df['Function Name'].fillna("").astype(str)

# Now your original check will work without errors
index = 0  # Example index
if df['Function Name'][index].find('PowerMode') != -1:
    print("Found PowerMode!")
else:
    print("PowerMode not found.")

Solution 2: Check type before calling find() (no column modification) #

If you don't want to alter the original column type, add a type check before using find():

index = 0
cell_value = df['Function Name'][index]
if isinstance(cell_value, str) and cell_value.find('PowerMode') != -1:
    print("Found PowerMode!")
else:
    print("PowerMode not found or cell is not a string.")

As a bonus, using pandas' vectorized operations instead of looping through indexes is more efficient. For example:

# Filter rows where Function Name contains "PowerMode" (handles NaN automatically)
matching_rows = df[df['Function Name'].fillna("").str.contains('PowerMode')]

内容的提问来源于stack exchange，提问作者Agar