非唯一索引上的 Dataframe.loc

Question

I have a dataframe with a non-unique index. I want to use .loc on a dataframe .

data = [['tom', 10], ['nick', 15], ['juli', 14], ['tom', 12], ['tom', 64]]
df = pd.DataFrame(data, columns = ['Name', 'Age'])
df.set_index('Name', inplace = True)

When I try

df.loc['tom': 'Age']

KeyError: "Cannot get left slice bound for non-unique label: 'tom'"

Why is that? Can I only .loc on dataframes with unique indexes?

Answer 1

You implementation is wrong. Age is not an element of Name index and moreover if you want to slice between rows then rows must be unique. Suppose you can't apply .loc[] on Name index like 'tom':'nick' because Name are not unique.

Check pandas.loc[] for more details.

You could do something like this:

import pandas as pd


data = [['tom', 10], ['nick', 15], ['juli', 14], ['tom', 12], ['tom', 64]]
df = pd.DataFrame(data, columns = ['Name', 'Age'])
df.set_index('Name', inplace = True)
df = df.loc['tom', 'Age']

print(df)

Output:

Name
tom    10
tom    12
tom    64
Name: Age, dtype: int64