如何对中文部分进行独热处理（one-hot）

import pandas as pd 
import numpy as np 

a = [['苹果', 10, 4.2], ['龙眼', '6.2,6.8', 0.03], ['荔枝', np.nan, np.nan]]
df = pd.DataFrame(a)
df.columns=['a','b','c']
print(df)

from sklearn import preprocessing  
enc = preprocessing.OneHotEncoder()  
# df['a'].encode('Unicode')
enc.fit(df['a'])  
df['a'] = enc.transform([df['a']])
print('df[a]\n',df['a'])

输出结果：

    a        b     c
0  苹果       10  4.20
1  龙眼  6.2,6.8  0.03
2  荔枝      NaN   NaN

报错：

ValueError: could not convert string to float: '荔枝'