如何截取Pandas DataFrame首尾？用tabulate展示首尾5行数据

Hey there! Let's break down your two questions about Pandas and tabulate:

Pandas makes this super straightforward with two built-in methods:

• head(n): Grabs the first n rows of your DataFrame. If you don't specify n, it defaults to 5. For example, df.head(10) gives you the first 10 rows.
• tail(n): Fetches the last n rows, also defaulting to 5 if n isn't provided. df.tail(10) gets you the final 10 rows.

If you want to combine both the start and end sections into a single DataFrame, just use pd.concat() to stitch them together:

import pandas as pd

# Example 100-row DataFrame
df = pd.DataFrame({'ColumnA': range(100), 'ColumnB': range(100, 200)})

# Combine first 5 and last 5 rows
trimmed_df = pd.concat([df.head(5), df.tail(5)], ignore_index=True)
print(trimmed_df)

To get that clean truncated table with an ellipsis in the middle, we can build a custom DataFrame that includes your top rows, a row of ellipses, and your bottom rows, then pass that to tabulate. Here's how:

import pandas as pd
from tabulate import tabulate

# Your original 100-row DataFrame (replace with your actual data)
df = pd.DataFrame({
    'Col1': range(100),
    'Col2': [x*2 for x in range(100)],
    'Col3': [x+50 for x in range(100)]
})

# Grab the top 5 and bottom 5 rows
top_rows = df.head(5)
bottom_rows = df.tail(5)

# Create a row of ellipses that matches your DataFrame's columns
ellipsis_row = pd.DataFrame([['...'] * len(df.columns)], columns=df.columns)

# Combine all three parts
display_df = pd.concat([top_rows, ellipsis_row, bottom_rows], ignore_index=True)

# Print with tabulate using your preferred format
print(tabulate(display_df, headers='keys', tablefmt='psql'))

This will output a nice psql-style table that looks like this:

+----+-------+-------+-------+
|    | Col1  | Col2  | Col3  |
|----+-------+-------+-------|
|  0 | 0     | 0     | 50    |
|  1 | 1     | 2     | 51    |
|  2 | 2     | 4     | 52    |
|  3 | 3     | 6     | 53    |
|  4 | 4     | 8     | 54    |
|  5 | ...   | ...   | ...   |
|  6 | 95    | 190   | 145   |
|  7 | 96    | 192   | 146   |
|  8 | 97    | 194   | 147   |
|  9 | 98    | 196   | 148   |
| 10 | 99    | 198   | 149   |
+----+-------+-------+-------+

The key here is making sure the ellipsis row uses the exact same column names as your original DataFrame—this ensures tabulate renders the table correctly with matching headers.

内容的提问来源于stack exchange，提问作者ferengi