利用python进行数据分析笔记.pdf

Python import pandas as pd import numpy as np import matplotlib.pyplot as plt import statsmodels.api as sm from pandas import DataFrame,Series CH02 JSON records = [json.loads(line) for line in open(path)] time_zone = [rec[tz] for rec in records if tz in rec] 0 from collections import defaultdict counts = defaultdict(int) top10 from collections import Counter toptz = Counter(time_zone) toptz.most_common(10) pandas from pandas import DataFrame,Series frame = DataFrame(records) tz_counts = frame[tz].value_counts() // tz_counts[:10] //top10 #####fillnaNA clean_tz = frame[tz].fillna(Missing) clean_tz[clean_tz == ] = ‘Unknown’ // clean_tz== #####Seriesplot tz_counts[:10].plot(kind=barh,rot=0) // type(tz_counts) //pandas.core.series.Series #####DataFramenull na frame.a.notnull() //True False clean_frame = frame[frame.a.notnull()]frame.a.dropna() MovieLens users = pd.read_table(xx/movielens/users.dat,sep=::,header=None,names= [userid,gender,age,occupation,zip’]) //:: ratings = pd.read_table(./Downloads/python/pydata- book-master/ch02/movielens/ratings.dat,sep=::,header=None,names= [userid,movieid,rating,timestamp]) movies = pd.read_table(./Downloads/python/pydata- book-master/ch02/movielens/movies.dat,sep=::,header=None,names= [movieid,title,genres]) #####pandasmergeusersratings data = pd.merge(pd.merge(ratings,users),movies) //cool! data.ix[0] // #####pandaspivot_table mean_ratings = data.pivot_table(rating,index=title,columns=gender,aggfunc=m ean’) // #####250 rating_by_title = data.groupby(title).size() // title $1,000,000 Duck (1971) 37 Night Mother (1986) 70 Til There Was You (1997) 52 active_titles = rating_by_title.index[rating_by_title = 250] // mean_ratings = mean_ratings.ix[acti