import pandas as pd
from io import StringIO

csv_data = '''A,B,C,D
1,2,3,4
5,6,,8
0,11,12,'''

df = pd.read_csv(StringIO(csv_data))
print(df)
#統(tǒng)計為空的數(shù)目
print(df.isnull().sum())
print(df.values)

#丟棄空的
print(df.dropna())
print('after', df)
from sklearn.preprocessing import Imputer
# axis=0 列  axis = 1 行
imr = Imputer(missing_values='NaN', strategy='mean', axis=0)
imr.fit(df) # fit 構(gòu)建得到數(shù)據(jù)
imputed_data = imr.transform(df.values) #transform 將數(shù)據(jù)進行填充
print(imputed_data)

df = pd.DataFrame([['green', 'M', 10.1, 'class1'],
          ['red', 'L', 13.5, 'class2'],
          ['blue', 'XL', 15.3, 'class1']])
df.columns =['color', 'size', 'price', 'classlabel']
print(df)

size_mapping = {'XL':3, 'L':2, 'M':1}
df['size'] = df['size'].map(size_mapping)
print(df)

## 遍歷Series
for idx, label in enumerate(df['classlabel']):
  print(idx, label)

#1, 利用LabelEncoder類快速編碼,但此時對color并不適合,
#看起來,好像是有大小的
from sklearn.preprocessing import LabelEncoder
class_le = LabelEncoder()
color_le = LabelEncoder()
df['classlabel'] = class_le.fit_transform(df['classlabel'].values)
#df['color'] = color_le.fit_transform(df['color'].values)
print(df)

#2, 映射字典將類標轉(zhuǎn)換為整數(shù)
import numpy as np
class_mapping = {label: idx for idx, label in enumerate(np.unique(df['classlabel']))}
df['classlabel'] = df['classlabel'].map(class_mapping)
print('2,', df)


#3,處理1不適用的
#利用創(chuàng)建一個新的虛擬特征
from sklearn.preprocessing import OneHotEncoder
pf = pd.get_dummies(df[['color']])
df = pd.concat([df, pf], axis=1)
df.drop(['color'], axis=1, inplace=True)
print(df)