数据分析：1880 —— 2020全美婴儿姓名

# -*- coding: utf-8 -*-

import pandas as pd
import matplotlib.pyplot as plt
import numpy as np

names1880 = pd.read_csv('names/yob1880.txt', names=[ 'name', 'sex', 'births' ])
#print names1880

#用birth列的six分组
#print names1880.groupby('sex').births.sum()

#组装数据并加上年限
years = range(1880, 2011)
pieces = []
columns = [ 'name', 'sex', 'births' ]
for year in years:
    path = 'names/yob%d.txt' % year
    frame = pd.read_csv(path, names=columns)
    
    frame[ 'year' ] = year
    pieces.append( frame )

#将所有数据整合到DataFrame中
#ignore_index=True去掉原始行号
names = pd.concat( pieces, ignore_index=True)
#print names[5000:5010]

#聚合
#.tail() 返回后几行，默认等于5
total_births = names.pivot_table( 'births', rows='year', cols='sex', aggfunc=sum )
#print total_births.tail()

#婴儿占总出生数的比例
def add_prop(group):