python 学习记录

对照着Python for data analysis学的，这本书真心好。学了几页就可以画出简单的图形了。强烈推荐

##---(Mon Nov 21 22:42:40 2016)---import pandaspath = "D:\1py\pydata-book-master\ch02\usagov_bitly_data2012-03-16-1331923249.txt"path = "D:/1py/pydata-book-master/ch02/usagov_bitly_data2012-03-16-1331923249.txt"open(path).readline()import jsonrecords = [json.load(line) for line in open(path)]records = [json.loads(line) for line in open(path)]resords[0]records[0]records[0]['u']print(records[0]['u'])times_zones = [rec['tz'] for rec in records]times_zones = [rec['tz'] for rec in records if 'tz' in rec]times_zones[:10]def get_counts(sequence):    counts = {}    for x in sequence:        if x in counts:            counts[x] + =1def get_counts(sequence):    counts = {}    for x in sequence:        if x in counts:            counts[x] + = 1def get_counts(sequence):    counts = {}    for x in sequence:        if x in counts:            counts[x] =counts[x] + 1        else:            counts[x] = 1    return countscounts = get_counts(time_zones)counts = get_counts(times_zones)counts["America/NewYork"]counts['America/NewYork']counts['America/New_York']len(times_zones)def get_counts2(squence):    counts = defaultdic(int)    for x in squence:        counts[x] = counts[x]+1    return countscounts = get_counts2(times_zones)from collections import defaultdictcounts = get_counts2(times_zones)def get_counts2(squence):    counts = defaultdict(int)    for x in squence:        counts[x] = counts[x]+1    return countscounts = get_counts2(times_zones)counts['America/New_York']def top_counts(count_dic, n=10):    value_key_pairs = [(count, tz) for tz ,count in count_dict.items()]    value_key_pairs.sort()    return value_key_pairs[-n:]top_counts(counts)def top_counts(count_dict, n=10):    value_key_pairs = [(count, tz) for tz ,count in count_dict.items()]    value_key_pairs.sort()    return value_key_pairs[-n:]top_counts(counts)from pandas imort DataFrame, Seriesfrom pandas import DataFrame, Seriesimport pandas as pdframe = DataFrame(records)frameframe['tz'][:10]tz_counts = frame['tz'].value_counts()tz_counts[:10]clean_tz = frame['tz'].fillna("Missing")clean_tz[clean_tz  == ''] ="Unknown"tz_counts = clean_tz.value_counts()tz_counts[:10]tz_counts[:10].plot(kind="barh", rot=0)