五一出行你准备好了吗？Python帮你预测全国热门景点！

小长假第一天总是最堵的

五一自驾游 西湖成最热门首选

# 导入包import requestsimport jsonimport pandas as pdimport timedef get_qunaer_one_city(search_city):    """    功能：给定城市和页数，获取一页的信息    """    # 获取URL    url_1 = 'https://piao.qunar.com/ticket/list.json?keyword={}&page=1'.format(search_city)    # 添加Headers    headers = {        'user-agent': 'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/55.0.2883.87 Safari/537.36'    }    # 先发起一次请求    response = requests.get(url_1, headers=headers)    data = response.json()['data']    # 所有页数    totalCount = int(data.get('totalCount') / 15)+1    # 存储数据    df_all = pd.DataFrame()    # 循环获取    for page_num in range(1, totalCount+1):        url_2 = 'https://piao.qunar.com/ticket/list.json?keyword={}&page={}'.format(search_city, page_num)        # 发起请求        try:            response = requests.get(url_2, headers=headers, timeout=3)        except ProxyError:            response = requests.get(url_2, headers=headers, timeout=3)        # 获取数据        data = response.json()['data']        # 景点名称        sightName = [i.get('sightName') for i in data['sightList']]        # 景点星级        star = [i.get('star', '无') for i in data['sightList']]        # 景点评分        score = [i.get('score', 0) for i in data['sightList']]        # 景点介绍        intro = [i.get('intro', '') for i in data['sightList']]        # 景点经纬度        point = [i.get('point') for i in data['sightList']]        # 景点地区        districts = [i.get('districts') for i in data['sightList']]        # 相关景点        childrenCount = [i.get('childrenCount', 0) for i in data['sightList']]        # 月销量        saleCount = [i.get('saleCount', 0) for i in data['sightList']]        # 门票价格        Price = [i.get('qunarPrice', 0) for i in data['sightList']]        # 保存数据        df_one = pd.DataFrame({            'sightName': sightName,            'star': star,            'score': score,            'intro': intro,            'point': point,            'districts': districts,            'childrenCount': childrenCount,            'saleCount': saleCount,            'qunarPrice': Price        })        # 追加数据        df_all = df_all.append(df_one, ignore_index=True)        # 打印进度        print('我正在获取第{}页'.format(page_num))        # 休眠一秒        time.sleep(1)    return df_all

df_all.head()

# 导入包import numpy as npimport pandas as pd import os from pyecharts.charts import Bar, Pie, Map, Page, BMap, Scatterfrom pyecharts import options as opts 

# 获取文件路径data_files = os.listdir('./data/')# 从磁盘中循环读取数据df_all = pd.DataFrame()for file in data_files:    df_one = pd.read_excel('./data/{}'.format(file))    df_all = df_all.append(df_one, ignore_index=True)# 复制一份df_qa = df_all.copy()# 删除重复值df_qa.drop_duplicates(inplace=True)# 提取省份df_qa['province'] = df_qa.districts.str.split('·').apply(lambda x:x[0]) # 计算销售额df_qa['sales_volume'] = df_qa['saleCount'] * df_qa['qunarPrice']

# 景点星级分布star_num = df_qa['star'].value_counts() # 数据对data_pair = [list(z) for z in zip(star_num.index, star_num.values.tolist())]# 饼图pie1 = Pie(init_opts=opts.InitOpts(width='1350px', height='750px'))pie1.add(         series_name="num",        radius=["35%", "60%"],        data_pair=data_pair,        label_opts=opts.LabelOpts(            position="outside",            formatter="{a|{a}}{abg|}\n{hr|}\n {b|{b}: }{c}  {per|{d}%}  ",            background_color="#eee",            border_color="#aaa",            border_width=1,            border_radius=4,            rich={                "a": {"color": "#999", "lineHeight": 22, "align": "center"},                "abg": {                    "backgroundColor": "#e3e3e3",                    "width": "100%",                    "align": "right",                    "height": 22,                    "borderRadius": [4, 4, 0, 0],                },                "hr": {                    "borderColor": "#aaa",                    "width": "100%",                    "borderWidth": 0.5,                    "height": 0,                },                "b": {"fontSize": 16, "lineHeight": 33},                "per": {                    "color": "#eee",                    "backgroundColor": "#334455",                    "padding": [2, 4],                    "borderRadius": 2,                },            },        ),)pie1.set_global_opts(legend_opts=opts.LegendOpts(pos_left="left", pos_top='30%', orient="vertical"),                      title_opts=opts.TitleOpts(title='全国景点星级分布'))pie1.set_series_opts(    tooltip_opts=opts.TooltipOpts(trigger="item", formatter="{a} <br/>{b}: {c} ({d}%)")    )pie1.render() 

# 国内城市top15city_top10 = df_qa.province.value_counts()[:15]# 条形图bar1 = Bar(init_opts=opts.InitOpts(width='1350px', height='750px'))bar1.add_xaxis(city_top10.index.tolist())bar1.add_yaxis("省份", city_top10.values.tolist())bar1.set_global_opts(title_opts=opts.TitleOpts(title="全国景点数量Top10"),                     visualmap_opts=opts.VisualMapOpts(max_=1200))bar1.render() 

saleCount_num = df_qa.groupby('province')['saleCount'].sum().sort_values(ascending=False) # 地图map1 = Map(init_opts=opts.InitOpts(width='1350px', height='750px'))map1.add("", [list(z) for z in zip(saleCount_num.index.tolist(), saleCount_num.values.tolist())],         maptype='china') map1.set_global_opts(title_opts=opts.TitleOpts(title='各省份销量热力图'),                     visualmap_opts=opts.VisualMapOpts(max_=80000))map1.render() 

推荐阅读 READ MORE