在Pandas中实现DataFrame行乘法运算（无需转数组）

Got it, let's tackle this without converting your DataFrame to arrays—Pandas has built-in functions that handle this cleanly, including skipping null values automatically.

Here's a step-by-step approach that keeps everything within Pandas:

1. Handle Decimal Separators (If Needed) #

First, notice your sample data uses commas as decimal separators (like 0,03). Pandas expects dots for numeric values, so we'll convert those first if your actual data is stored as strings:

import pandas as pd

# Replace commas with dots and convert to float for all numeric columns
numeric_cols = df.select_dtypes(exclude=['object']).columns  # Or list your cols explicitly if needed
df[numeric_cols] = df[numeric_cols].replace(',', '.', regex=True).astype(float)

2. Calculate the Row-wise Product #

We'll select all numeric columns, add 1 to each value, then compute the product across each row—with skipna=True to ignore null values:

# Select numeric columns (excludes the 'date' column automatically if it's a datetime/object type)
numeric_data = df.select_dtypes(include=['float64', 'int64'])

# Compute (1 + value) product per row, skipping nulls
row_product = (1 + numeric_data).prod(axis=1, skipna=True)

3. Build the New DataFrame #

Finally, construct your desired output by combining the original date column with the calculated product values:

new_df = pd.DataFrame({
    'date': df['date'],
    'value': row_product
})

Why This Works #

• No array conversion required: All operations use Pandas' native DataFrame/Series methods, so you won't lose any data or metadata.
• skipna=True (default for prod()) ensures any null cells in the numeric columns are skipped during the product calculation.
• select_dtypes makes this flexible—if you add more numeric columns later (like F, G), the code will automatically include them without modification.

内容的提问来源于stack exchange，提问作者khouzam