Pandas 学习笔记

• 遍历DataFrame

method 1

for index, row in dataframe.iterrows():
    row['column_name']

method 2 

for row in dataframe.itertuples(index=True, name='Pandas'):
    getattr(row,'column_name')

method 3 

for i in range(len(dataframe)):
    dataframe.iloc[i]['column_name]

• 特征离散化处理(id 类) : map 映射

import pandas

data = pandas.DataFrame([['id1','class1'],['id2','class2'],['id3','class1']])
data.columns = ['id','class']

print('previous data :\n',data)

mapping = {'id1':1,'id2':2,'id3':3}
# 映射字典

data['id'] = data['id'].map(mapping)

<