ch09-GroupBy
split->apply->combine 12 import numpy as npimport pandas as pd 12345 df = pd.DataFrame({'key1' : ['a', 'a', 'b', 'b', 'a'], 'key2' : ['one', 'two', 'one', 'two', 'one'], 'data1' : np.random.randn(5), 'data2' : np.random.randn(5)})df */ /*--> */ key1 key2 data1 data2 0 a one 1.587125 -0.517650 1 a two 0.206854 1.503013 2 b one 1.074688 -1.310088 3 b two 0.306591 2.236456 4 a one 0.462624 0.643336 123 grouped = df['data1'].groupby(df['key1'])grouped #一个groupby对象 <pandas.core.groupby.groupby.SeriesGroupBy object at 0x10637ba20> 1 grouped.mean() key1 a 0.752201 b 0.690639 Name: data1, dtype: