数据聚合与分组处理——数据聚合、分组运算与转换、透视表与交叉表
代码
# -*- coding: utf-8 -*-
from __future__ import division
from numpy.random import randn
import numpy as np
import os
import matplotlib.pyplot as plt
np.random.seed(12345)
plt.rc('figure', figsize=(10, 6))
from pandas import Series, DataFrame
import pandas as pd
np.set_printoptions(precision=4)
pd.options.display.notebook_repr_html = False
get_ipython().magic(u'matplotlib inline')
### GroupBy 技术
df = DataFrame({'key1' : ['a', 'a', 'b', 'b', 'a'],
'key2' : ['one', 'two', 'one', 'two', 'one'],
'data1' : np.random.randn(5),
'data2' : np.random.randn(5)})
df
grouped = df['data1'].groupby(df['key1'])
grouped
grouped.mean()
means = df['data1'].groupby([df['key1'], df['key2']]).mean()
means
means.unstack()
states = np.array(['Ohio', 'California', 'California', 'Ohio', 'Ohio'])
years = np.array([2005, 2005, 2006, 2005, 2006])
df['data1'].groupby([states, years]).mean()
df.groupby('key1').mean()
df.groupby(['key1', 'key2']).mean()
df.groupby(['key1', 'key2']).size()
# ### 对分组进行迭代
for name, group in df.groupby('key1'):
print(name)
print(group)
df.groupby('key1')
for (k1, k2), group in df.groupby(['key1', 'key2']):
print((k1, k2))
print(group)
pieces = dict(list(df.groupby('key1')))
pieces['b']
df.dtypes
grouped = df.groupby(df.dtypes, axis=1)
dict(list(grouped))
# ### 选择一个或一组列
df.groupby('key1')['data1']
df.groupby('key1')[['data2']]
df['data1'].groupby(df['key1'])
df[['data2']].groupby(df['key1'])
df.groupby(['key1', 'key2'])[['data2']].mean()
s_grouped = df.groupby(['key1', 'key2'])['data2']
s_grouped
s_grouped.mean()
# ### 通过字典或series进行分组
people = DataFrame(np.random.randn(5, 5),
columns=['a', 'b', 'c', 'd', 'e'],
index=['Joe', 'Steve', 'Wes', 'Jim', 'Travis'])
people.ix[2:3, ['b', 'c']] = np.nan # Add a few NA values
people
mapping = {'a': 'red', 'b': 'red', 'c': 'blue',
'd': 'blue', 'e': 'red', 'f' : 'orange'}
by_column = people.groupby(mapping, axis=1)
by_column.sum()
map_series = Series(mapping)
map_series
people.groupby(map_series, axis=1).count()
# ### 通过函数进行分组
people.groupby(len).sum()
key_list = ['one', 'one', 'one', 'two', 'two']
people.groupby([len, key_list]).min()
# ### 通过索引进行分组
columns = pd.MultiIndex.from_arrays([['US', 'US', 'US', 'JP', 'JP'],
[1, 3, 5, 1, 3]], names=['cty', 'tenor'])
hier_df = DataFrame(np.random.randn(4, 5), columns=columns)
hier_df
hier_df.groupby(level='cty', axis=1).count()
# ##数据聚合
df
grouped = df.groupby('key1')
grouped['data1'].quantile(0.9)
def peak_to_peak(arr):
return arr.max() - arr.min()
grouped.agg(peak_to_peak)
grouped.describe()
# ### 面向列的多函数应用
tips = pd.read_csv('d:/data/tips.csv')
tips['tip_pct'] = tips['tip'] / tips['total_bill']
tips[:6]
grouped = tips.groupby(['sex', 'smoker'])
grouped_pct = grouped['tip_pct']
grouped_pct.agg('mean')
grouped_pct.agg(['mean', 'std', peak_to_peak])
grouped_pct.agg([('foo', 'mean'), ('bar', np.std)])
functions = ['count', 'mean', 'max']
result = grouped['tip_pct', 'total_bill'].agg(functions)
result
result['tip_pct']
ftuples = [('Durchschnitt', 'mean'), ('Abweichung', np.var)]
grouped['tip_pct', 'total_bill'].agg(ftuples)
grouped.agg({'tip' : np.max, 'size' : 'sum'})
grouped.agg({'tip_pct' : ['min', 'max', 'mean', 'std'],
'size' : 'sum'})
# ##分组级运算和转换
df
k1_means = df.groupby('key1').mean().add_prefix('mean_')
k1_means
pd.merge(df, k1_means, left_on='key1', right_index=True)
people
key = ['one', 'two', 'one', 'two', 'one']
people.groupby(key).mean()
people.groupby(key).transform(np.mean)
def demean(arr):
return arr - arr.mean()
demeaned = people.groupby(key).transform(demean)
demeaned
demeaned.groupby(key).mean()
# ### apply方法
def top(df, n=5, column='tip_pct'):
return df.sort_index(by=column)[-n:]
top(tips, n=6)
tips.groupby('smoker').apply(top)
tips.groupby(['smoker', 'day']).apply(top, n=1, column='total_bill')
result = tips.groupby('smoker')['tip_pct'].describe()
result
result.unstack('smoker')
#f = lambda x: x.describe()
#grouped.apply(f)
# 禁止分组键
tips.groupby('smoker', group_keys=False).apply(top)
# ### 分位数和桶分析
frame = DataFrame({'data1': np.random.randn(1000),
'data2': np.random.randn(1000)})
factor = pd.cut(frame.data1, 4)
factor[:10]
def get_stats(group):
return {'min': group.min(), 'max': group.max(),
'count': group.count(), 'mean': group.mean()}
grouped = frame.data2.groupby(factor)
grouped.apply(get_stats).unstack()
grouping = pd.qcut(frame.data1, 10, labels=False)
grouped = frame.data2.groupby(grouping)
grouped.apply(get_stats).unstack()
# ### 用特定于分组的值填充缺失值
s = Series(np.random.randn(6))
s[::2] = np.nan
s
s.fillna(s.mean())
states = ['Ohio', 'New York', 'Vermont', 'Florida',
'Oregon', 'Nevada', 'California', 'Idaho']
group_key = ['East'] * 4 + ['West'] * 4
data = Series(np.random.randn(8), index=states)
data[['Vermont', 'Nevada', 'Idaho']] = np.nan
data
data.groupby(group_key).mean()
fill_mean = lambda g: g.fillna(g.mean())
data.groupby(group_key).apply(fill_mean)
fill_values = {'East': 0.5, 'West': -1}
fill_func = lambda g: g.fillna(fill_values[g.name])
data.groupby(group_key).apply(fill_func)
# ### 随机采样和排列
suits = ['H', 'S', 'C', 'D']
card_val = (range(1, 11) + [10] * 3) * 4
base_names = ['A'] + range(2, 11) + ['J', 'K', 'Q']
cards = []
for suit in ['H', 'S', 'C', 'D']:
cards.extend(str(num) + suit for num in base_names)
deck = Series(card_val, index=cards)
deck[:13]
def draw(deck, n=5):
return deck.take(np.random.permutation(len(deck))[:n])
draw(deck)
get_suit = lambda card: card[-1] #只要最后一个字母
deck.groupby(get_suit).apply(draw, n=2)
#不显示分组关键字
deck.groupby(get_suit, group_keys=False).apply(draw, n=2)
# ### 分组加权平均数和相关系数
df = DataFrame({'category': ['a', 'a', 'a', 'a', 'b', 'b', 'b', 'b'],
'data': np.random.randn(8),
'weights': np.random.rand(8)})
df
grouped = df.groupby('category')
get_wavg = lambda g: np.average(g['data'], weights=g['weights'])
grouped.apply(get_wavg)
close_px = pd.read_csv('d:/data/stock_px.csv', parse_dates=True, index_col=0)
close_px.info()
close_px[-4:]
rets = close_px.pct_change().dropna()
spx_corr = lambda x: x.corrwith(x['SPX'])
by_year = rets.groupby(lambda x: x.year)
by_year.apply(spx_corr)
# 苹果公司和微软的年度相关系数
by_year.apply(lambda g: g['AAPL'].corr(g['MSFT']))
# ## 透视表
tips.pivot_table(index=['sex', 'smoker'])
tips.pivot_table(['tip_pct', 'size'], index=['sex', 'day'],
columns='smoker')
tips.pivot_table(['tip_pct', 'size'], index=['sex', 'day'],
columns='smoker', margins=True)
tips.pivot_table('tip_pct', index=['sex', 'smoker'], columns='day',
aggfunc=len, margins=True)
tips.pivot_table('size', index=['time', 'sex', 'smoker'],
columns='day', aggfunc='sum', fill_value=0)
# ### 交叉表
from StringIO import StringIO
data = """Sample Gender Handedness
1 Female Right-handed
2 Male Left-handed
3 Female Right-handed
4 Male Right-handed
5 Male Left-handed
6 Male Right-handed
7 Female Right-handed
8 Female Left-handed
9 Male Right-handed
10 Female Right-handed"""
data = pd.read_table(StringIO(data), sep='\s+')
data
pd.crosstab(data.Gender, data.Handedness, margins=True)
pd.crosstab([tips.time, tips.day], tips.smoker, margins=True)
# ## 2012联邦选举委员会数据分析
fec = pd.read_csv('d:/data/P00000001-ALL.csv')
fec.info()
fec.ix[123456]
unique_cands = fec.cand_nm.unique()
unique_cands
unique_cands[2]
parties = {'Bachmann, Michelle': 'Republican',
'Cain, Herman': 'Republican',
'Gingrich, Newt': 'Republican',
'Huntsman, Jon': 'Republican',
'Johnson, Gary Earl': 'Republican',
'McCotter, Thaddeus G': 'Republican',
'Obama, Barack': 'Democrat',
'Paul, Ron': 'Republican',
'Pawlenty, Timothy': 'Republican',
'Perry, Rick': 'Republican',
"Roemer, Charles E. 'Buddy' III": 'Republican',
'Romney, Mitt': 'Republican',
'Santorum, Rick': 'Republican'}
fec.cand_nm[123456:123461]
fec.cand_nm[123456:123461].map(parties)
fec['party'] = fec.cand_nm.map(parties)
fec['party'].value_counts()
(fec.contb_receipt_amt > 0).value_counts()
fec = fec[fec.contb_receipt_amt > 0]
fec_mrbo = fec[fec.cand_nm.isin(['Obama, Barack', 'Romney, Mitt'])]
# #根据职业和雇主统计赞助信息
fec.contbr_occupation.value_counts()[:10]
occ_mapping = {
'INFORMATION REQUESTED PER BEST EFFORTS' : 'NOT PROVIDED',
'INFORMATION REQUESTED' : 'NOT PROVIDED',
'INFORMATION REQUESTED (BEST EFFORTS)' : 'NOT PROVIDED',
'C.E.O.': 'CEO'
}
# If no mapping provided, return x
f = lambda x: occ_mapping.get(x, x)
fec.contbr_occupation = fec.contbr_occupation.map(f)
emp_mapping = {
'INFORMATION REQUESTED PER BEST EFFORTS' : 'NOT PROVIDED',
'INFORMATION REQUESTED' : 'NOT PROVIDED',
'SELF' : 'SELF-EMPLOYED',
'SELF EMPLOYED' : 'SELF-EMPLOYED',
}
# If no mapping provided, return x
f = lambda x: emp_mapping.get(x, x)
fec.contbr_employer = fec.contbr_employer.map(f)
by_occupation = fec.pivot_table('contb_receipt_amt',
index='contbr_occupation',
columns='party', aggfunc='sum')
over_2mm = by_occupation[by_occupation.sum(1) > 2000000]
over_2mm
over_2mm.plot(kind='barh')
def get_top_amounts(group, key, n=5):
totals = group.groupby(key)['contb_receipt_amt'].sum()
# Order totals by key in descending order
return totals.order(ascending=False)[-n:]
grouped = fec_mrbo.groupby('cand_nm')
grouped.apply(get_top_amounts, 'contbr_occupation', n=7)
grouped.apply(get_top_amounts, 'contbr_employer', n=10)
# #对出资额分组
bins = np.array([0, 1, 10, 100, 1000, 10000, 100000, 1000000, 10000000])
labels = pd.cut(fec_mrbo.contb_receipt_amt, bins)
labels
grouped = fec_mrbo.groupby(['cand_nm', labels])
grouped.size().unstack(0)
bucket_sums = grouped.contb_receipt_amt.sum().unstack(0)
bucket_sums
normed_sums = bucket_sums.div(bucket_sums.sum(axis=1), axis=0)
normed_sums
normed_sums[:-2].plot(kind='barh', stacked=True)
# #根据州统计赞助信息
grouped = fec_mrbo.groupby(['cand_nm', 'contbr_st'])
totals = grouped.contb_receipt_amt.sum().unstack(0).fillna(0)
totals = totals[totals.sum(1) > 100000]
totals[:10]
percent = totals.div(totals.sum(1), axis=0)
percent[:10]
