擅长:python、mysql、java
<p>您可以使用<a href="https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.Categorical.html" rel="nofollow noreferrer">^{<cd1>}</a>来处理这些类型的问题<code>categorical</code>类型还有一个额外的好处,就是内存效率更高</p>
<h3>示例:</h3>
<pre><code>cars = ['Ford', 'Mercedes-Benz', 'Nissan', 'Honda', 'Renault', 'BMW',
'Land Rover', 'Volkswagen', 'Audi', 'Chrysler', 'Jaguar',
'Mitsubishi', 'Kia', 'Porsche', 'Toyota', 'Hyundai']
df = pd.DataFrame({'cars': np.random.choice(cars, 100)})
top_5 = ['Ford', 'Mercedes-Benz', 'Nissan', 'Audi', 'Jaguar', 'Miscellaneous']
df['cars_refined'] = pd.Categorical(df['cars'], categories=top_5).fillna('Miscellaneous')
print(df.head(10))
cars cars_refined
0 Mercedes-Benz Mercedes-Benz
1 Mercedes-Benz Mercedes-Benz
2 Volkswagen Miscellaneous
3 Ford Ford
4 Mitsubishi Miscellaneous
5 Toyota Miscellaneous
6 Porsche Miscellaneous
7 Honda Miscellaneous
8 Kia Miscellaneous
9 Jaguar Jaguar
</code></pre>