如何从Pandas DataFrame中筛选包含至少一个非零值的列？

Hey there! I get your frustration—filtering rows is straightforward, but targeting columns with non-zero values can feel tricky at first. Let's fix this with your sample DataFrame.

First, let's recreate your data to work with:

import pandas as pd

data = {
    'Type': ['Seg0', 'Seg1'],
    'ADR': [0.0, 0.0],
    'ALE': [0.0, 0.0],
    'ALF': [0.0, 0.5],
    'AME': [0.0, 0.0]
}
df = pd.DataFrame(data)

The Simple One-Liner Solution #

You can use a combination of boolean masking and the .any() method to select only columns that have at least one non-zero value:

filtered_df = df.loc[:, (df != 0).any()]

Let's break this down:

• df != 0 creates a boolean DataFrame where each cell is True if the value isn't zero, False otherwise.
• .any() checks each column and returns True if there's at least one True (i.e., at least one non-zero value) in that column.
• df.loc[:, ...] uses that boolean result to slice the DataFrame, keeping only the columns that meet the condition.

What the Output Looks Like #

Running this code will give you exactly what you want:

Type  ALF
0  Seg0  0.0
1  Seg1  0.5

Alternative Approach: Drop All-Zero Columns #

If you prefer, you can also achieve the same result by dropping columns that are entirely zero:

filtered_df = df.drop(columns=df.columns[(df == 0).all()])

Here, (df == 0).all() identifies columns where every value is zero, and drop() removes them from the DataFrame.

Both methods work perfectly—pick whichever makes more sense to you!

内容的提问来源于stack exchange，提问作者user3476463