def plot_geo_station_num(re_dic, to_path):
dw = my_(config.MYSQL_BI_RW_ENV)
muchcols_table = 'data_center.dc_stations_muchcols_enc'
sqls = "select city scale,avg(latitude) avg_lat,avg(longitude) avg_lng,count(*) numbers from %s where name like '%%%s%%' group by city order by numbers desc" % (
muchcols_table, re_dic['name'])
count_station_ = dw.to_dataframe(sqls)
# count_station_['scale'] = count_station_['scale'].apply(lambda x: change_city(x, pro))
coords = {}
for i in range(len(count_station_)):
coords[count_station_.loc[i, 'scale']] = [count_station_.loc[i, 'avg_lng'], count_station_.loc[i, 'avg_lat']]
num_Max = count_station_['numbers'].max()
name_str = "油站名包含'%s':" % re_dic['name'] if re_dic['name'] else ''
page = pyecharts.Page()
style = pyecharts.Style(width=1300, height=860, background_color='#404a66', title_color="#eee", title_pos="center")
geo = pyecharts.Geo(name_str + "油站分布地图", **style.init_style)
geo.add("", count_station_['scale'], count_station_['numbers'], visual_text_color="#eee", is_legend_show=False,
symbol_size=12, is_visualmap=True, visual_range=[0, num_Max],
tooltip_formatter='{b}',
label_emphasis_textsize=15,
label_emphasis_pos='right', geo_cities_coords=coords)
page.add(geo)
chart2 = pyecharts.Bar(name_str + "油站分布柱状图", **style.init_style)
chart2.add("", count_station_['scale'][:25], count_station_['numbers'][:25], visual_range=[0, num_Max],
is_label_show=True,
is_visualmap=True, visual_text_color='#eee')
page.add(chart2)
page.render(to_path)
def plot_oil_station_num_(re_dic, to_path):
standard_province = ['安徽', '贵州', '澳门', '北京', '重庆', '福建', '福建', '甘肃', '广东', '广西', '广州', '海南', '河北', '黑龙江', '河南', '湖北', '湖南', '江苏',
'江西', '吉林', '辽宁', '内蒙古', '宁夏', '青海', '山东', '上海', '陕西', '山西', '四川', '台湾', '天津', '香港', '新疆', '西藏', '云南', '浙江']
dw = my_(config.MYSQL_BI_RW_ENV)
muchcols_table = 'data_center.dc_stations_muchcols_enc'
if re_dic['scale'] == 'china':
sqls = "select province scale,count(*) numbers from %s where name like '%%%s%%' group by province order by numbers desc" % (muchcols_table, re_dic['name'])
count_station_ = dw.to_dataframe(sqls)
count_station_['scale'] = count_station_['scale'].apply(lambda x: change_city(x, standard_province))
else:
re_dic['scale'] = re_dic['scale'].split('省')[0]
re_dic['scale'] = re_dic['scale'].split('市')[0]
sqls = "select city scale,count(*) numbers from %s where province like '%%%s%%' and name like '%%%s%%' group by city order by numbers desc" % (muchcols_table, re_dic['scale'], re_dic['name'])
count_station_ = dw.to_dataframe(sqls)
num_Max = count_station_['numbers'].max()
name_str = "油站名包含'%s':" %re_dic['name'] if re_dic['name'] else ''
page = pyecharts.Page()
style = pyecharts.Style(width=1300, height=860, background_color='#606a79', title_color="#eee", title_pos="center")
map1 = pyecharts.Map(name_str+"油站分布地图", **style.init_style)
map1.add("", count_station_['scale'], count_station_['numbers'], maptype=re_dic['scale'], visual_range=[0, num_Max], is_label_show=True,
is_visualmap=True, visual_text_color='#eee')
page.add(map1)
chart2 = pyecharts.Bar(name_str+"油站分布柱状图", **style.init_style)
chart2.add("", count_station_['scale'][:25], count_station_['numbers'][:25], maptype=re_dic['scale'], visual_range=[0, num_Max], is_label_show=True,
is_visualmap=True, visual_text_color='#eee')
page.add(chart2)
page.render(to_path)
def func():
page = pyecharts.Page() # step 1
# bar
attr = ["衬衫", "羊毛衫", "雪纺衫", "裤子", "高跟鞋", "袜子"]
v1 = [5, 20, 36, 10, 75, 90]
v2 = [10, 25, 8, 60, 20, 80]
bar = pyecharts.Bar("柱状图数据堆叠示例")
bar.add("商家A", attr, v1, is_stack=True)
bar.add("商家B", attr, v2, is_stack=True)
page.add(bar) # step 2
# scatter3D
import random
data = [[
random.randint(0, 100),
random.randint(0, 100),
random.randint(0, 100)
] for _ in range(80)]
range_color = [
'#313695', '#4575b4', '#74add1', '#abd9e9', '#e0f3f8', '#ffffbf',
'#fee090', '#fdae61', '#f46d43', '#d73027', '#a50026'
]
scatter3D = pyecharts.Scatter3D("3D 散点图示例", width=1200, height=600)
scatter3D.add("", data, is_visualmap=True, visual_range_color=range_color)
page.add(scatter3D) # step 2
page.render() # step 3
# scatter_geo
attr = ['11', '22']
data = ['1', '2']
scatter_geo = pyecharts.Geo("地理位置视图", width=1200, height=600)
scatter_geo.add("", attr, data)
page.add(scatter_geo) # step 2
page.render() # step 3
data = [["广州", "北京"], ["广州", "上海"]]
scatter_geo = pyecharts.GeoLines("地理位置视图", width=1200, height=600)
scatter_geo.add("上午航班",
data,
geo_normal_color='#ff00ff',
geo_effect_period=9,
geo_effect_color='#ff0000',
geo_effect_symbol='arrow')
data = [["广州", "南昌"], ["广州", "成都"]]
scatter_geo.add("下午航班",
data,
geo_effect_traillength=1,
geo_emphasis_color='#0000ff')
page.add(scatter_geo) # step 2
page.render() # step 3
def read_1688_all(cid, cat_name):
page = pyecharts.Page() # 实例化page类
bar = read_1688_index(cid, cat_name)
page.add(bar)
attr_list = mongo_client2.get_1688attr(cid)
# attrs = []
# for attr in attr_list:
# attrs.append(attr)
rows_by_type = defaultdict(list)
for attr in attr_list:
rows_by_type[attr['attr_type']].append(attr)
for attr_type in rows_by_type.items():
# print(attr_type[1]['attr_value'])
attr_value_index_group = []
attr_value_group = []
for attr in attr_type[1]:
attr_value_index_group.append(attr['attr_value_index'])
attr_value_group.append(attr['attr_value'])
# print(attr_type[0]+"\t"+attr['attr_value'])
# attr = ["衬衫", "羊毛衫", "雪纺衫", "裤子", "高跟鞋", "袜子"]
# v1 = [11, 12, 13, 10, 10, 10]
pie = pyecharts.Pie(attr_type[0], title_pos='50%')
pie.add(attr_type[0],
attr_value_group,
attr_value_index_group,
radius=[40, 75],
label_text_color=None,
is_label_show=True,
legend_orient='vertical',
legend_pos='left')
page.add(pie)
# print(rows_by_type)
# pie
# attr = ["衬衫", "羊毛衫", "雪纺衫", "裤子", "高跟鞋", "袜子"]
# v1 = [11, 12, 13, 10, 10, 10]
# pie = Pie("饼图-圆环图示例", title_pos='50%')
# pie.add("", attr, v1, radius=[40, 75], label_text_color=None,
# is_label_show=True, legend_orient='vertical', legend_pos='left')
# TODO 配置Grid类
# grid = Grid(height=720, width=1200) # 初始化，参数可传page_title,width,height
# grid.add(bar, grid_bottom="60%", grid_left="60%") # 添加要展示的图表，并设置显示位置
# # grid.add(line, grid_bottom="60%", grid_right="60%") # 添加要展示的图表，并设置显示位置
# grid.add(pie, grid_bottom="60%", grid_right="60%")
# grid.add(scatter, grid_top="60%", grid_left="60%") # 添加要展示的图表，并设置显示位置
# grid.add(es, grid_top="60%", grid_right="60%") # 添加要展示的图表，并设置显示位置
# grid.render("Grid_并行显示多张图表.html")
# return grid
return page
def plot_station_lostavg(re_dic, to_path):
re_dic = add_whereadd_redic(re_dic)
dic_station, dic_smarttime = get_station_detail_(re_dic)
re_dic['stname'] = '%s-%s' % (dic_station['city'], dic_station['stname'])
dw = my_(config.MYSQL_BI_RW_ENV)
sql_mer = """SELECT
sum(if(spr_id<3,value_c,0)) users_7,sum(if(spr_id<4,value_c,0)) users_15,sum(value_c*value_e) lostday_a,sum(value_c) users_a,batch_date
FROM profile.pr_stations_user_dis_ where mer_id = %(mer_id)s and tar_id = 30 %(whereadd)s group by batch_date""" % re_dic
sql_all = """SELECT
sum(if(spr_id<3,value_c,0)) users_7,sum(if(spr_id<4,value_c,0)) users_15,sum(value_c*value_e) lostday_a,sum(value_c) users_a,batch_date
FROM profile.pr_platforms_user_dis_ where mer_id = 0 and tar_id = 30 %(whereadd)s group by batch_date""" % re_dic
data_mer = dw.to_dataframe(sql_mer)
data_all = dw.to_dataframe(sql_all)
page = pyecharts.Page(page_title='流失分析',)
X = data_all['batch_date']
tem_dic = dict(label_text_size=12, is_label_show=False,
is_fill=True, area_opacity=0.08, line_width=2, is_smooth=True, mark_line=["average"]
, xaxis_interval=1, xaxis_rotate=45)
line1 = pyecharts.Line("平均流失天数趋势：", "%(stname)s(%(mer_id)s)" % re_dic, width=1500, height=600)
lostavg_mer = (data_mer['lostday_a'] / data_mer['users_a']).round(2)
lostavg_all = (data_all['lostday_a'] / data_all['users_a']).round(2)
line1.add("油站", X, lostavg_mer, **tem_dic)
line1.add("全平台", X, lostavg_all, **tem_dic)
page.add(line1)
rrate_mer = (data_mer['users_15'] / data_mer['users_a']).round(4)
rrate_all = (data_all['users_15'] / data_all['users_a']).round(4)
line2 = pyecharts.Line("流失15日内的用户占比：", "%(stname)s(%(mer_id)s)" % re_dic, width=1500, height=600)
conb_ = np.append(rrate_mer, rrate_all)
tem_dic.update({'yaxis_min': conb_.min() * 0.6})
lineadd_sted(line2, dic_smarttime, X, conb_.max() * 1.2)
line2.add("油站", X, rrate_mer, **tem_dic)
line2.add("全平台", X, rrate_all, **tem_dic)
page.add(line2)
rrate_mer = (data_mer['users_7'] / data_mer['users_a']).round(4)
rrate_all = (data_all['users_7'] / data_all['users_a']).round(4)
line2 = pyecharts.Line("流失7日内的用户占比：", "%(stname)s(%(mer_id)s)" % re_dic, width=1500, height=600)
conb_ = np.append(rrate_mer, rrate_all)
tem_dic.update({'yaxis_min': conb_.min() * 0.6})
lineadd_sted(line2, dic_smarttime, X, conb_.max() * 1.2)
line2.add("油站", X, rrate_mer, **tem_dic)
line2.add("全平台", X, rrate_all, **tem_dic)
page.add(line2)
page.render(to_path)
def create_bar():
user_list = JobsInfo.objects.all().values_list("job_title", "job_salary", "job_exp")
page = pe.Page()
name = ["服装"]
name2 = ["电器"]
x = list(user_list[0])
y = ["50", 200, 100]
y2 = ["50", 300, 1400]
# bar
bar = pe.Bar("我的第一个图表", "这里是副标题", width=600, height=300)
bar.add(name, x, y)
bar.add(name2, x, y2)
page.add(bar)
# line
line = pe.Line('Demo Line', "这里是副标题", width=600, height=300)
line.add(name, x, y)
line.add(name2, x, y2)
page.add(line)
return page
def plot_geo_station(re_dic, to_path):
dw = my_(config.MYSQL_BI_RW_ENV)
muchcols_table = 'data_center.dc_stations_muchcols_enc'
sqls = "select name scale,latitude lat,longitude lng from %s where name like '%%%s%%'" % (muchcols_table, re_dic['name'])
count_station_ = dw.to_dataframe(sqls)
count_station_['numbers']=1
coords={}
for i in range(len(count_station_)):
coords[count_station_.loc[i,'scale']]=[count_station_.loc[i,'lng'],count_station_.loc[i,'lat']]
# num_Max = count_station_['numbers'].max()
name_str = "油站名包含'%s':" %re_dic['name'] if re_dic['name'] else ''
page = pyecharts.Page()
style = pyecharts.Style(width=1300, height=860, background_color='#404a49', title_color="#eee", title_pos="center")
geo = pyecharts.Geo(name_str+"油站分布地图",**style.init_style)
geo.add("",count_station_['scale'], count_station_['numbers'], visual_text_color="#eee", is_legend_show=False,
symbol_size=11, is_visualmap=True,
tooltip_formatter='{b}',
label_emphasis_textsize=15,
label_emphasis_pos='right',geo_cities_coords=coords)
page.add(geo)
page.render(to_path)
def Top10Holder(codenumber, startdate='2014-10-10'):
"""
Top10 holde 是将top10 股东用Pie 图进行可视化，其中包含一年股东变化的河流图.
Parameters
----------
codenumber: stock 代码
startdate='2014-10-10': 起始时间
Returns
-------
输出为 当前文件夹下 html ：top10.html
"""
if len(codenumber) != 6: #检查 code 长度 不符合6位全部错误
return False
elif len(codenumber) == 6:
datestr = startdate.split("-")
thisyear = datestr[0]
df2 = ts.top10_holders(code=codenumber, gdtype="1") # 下载数据
test = df2[1]["quarter"].tolist() #获取季度列表
df_ready = df2[1]
idxlist = []
for idx, val in enumerate(test):
a = val.split("-")
if a[0] == thisyear:
# print a[0],idx
idxlist.append(idx)
elif a[0] > thisyear:
idxlist.append(idx)
thing = df_ready.loc[idxlist]
thing = pd.DataFrame(thing)
thing = thing.sort_values(["quarter", "name"]) #排序所有流通股股东
quarterlist = thing.quarter.value_counts()
quarterlist = quarterlist.index.sort_values()
timeline = Timeline(
height=600,
width=800,
timeline_left=20,
timeline_right=100,
is_auto_play=False,
timeline_bottom=0,
timeline_symbol='diamond',
)
year_index = thing.quarter.apply(
lambda x: x[0:4]) # year list ('20xx')
yearlist = year_index.value_counts()
yearlist = yearlist.index.sort_values()
#print(quarterlist,year)
for quarters in list(quarterlist): # 将每个季度画成PIE 加入TIMELINE
df = thing[thing["quarter"] == quarters]
df.h_pro = df.h_pro.apply(lambda x: float(x))
df.hold = df.hold.apply(lambda x: float(x))
sum_pro = round(df.h_pro.sum(), 2)
sum_hold = round(np.mean(df.hold / df.h_pro) * sum_pro, 0)
s = pd.Series({
'name': '其他',
'h_pro': 100 - sum_pro,
'hold': sum_hold
})
df = df.append(s, ignore_index=True)
#df=df.sort_values(["name"])
pie = Pie('%s' % quarters + "-" + "top10holder",
title_pos='outside')
TopPie = pie.add(
'',
df.name,
df.hold,
radius=[20, 45],
center=[45, 50],
is_legend_show=True,
is_label_show=True,
# is_visualmap=True,
is_more_utils=True,
is_toolbox_show=True,
legend_pos='left',
legend_orient='vertical',
legend_top=25,
legend_text_size=12,
label_text_size=10,
label_text_color="#100",
tooltip_text_color="#F0F",
#is_label_emphasis=False,#是否显示高亮
label_formatter="{b}:\n {c}万股\n{d}%")
#gird= Grid()
#gird.add(pie,grid_bottom="20%")
#gird.add(tab,grid_bottom="60%")
#pietimeline.add(pie, quarters,)
timeline.add(
pie,
quarters,
)
#page = pyecharts.Page()
#page.add(pie)
#page.add(_treemap)
#timeline.add(pie,quarters,)
#timeline.add(_treemap,quarters,)
#pietimeline.top('40%')
trmtimeline = Timeline(
height=1200,
width=800,
timeline_left=20,
timeline_right=100,
is_auto_play=False,
timeline_bottom=0,
timeline_symbol='diamond',
)
#for y in list(yearlist):
datas_index = []
for i, _year in enumerate(year_index): #获取datas_index
if _year in list(yearlist)[-1:]:
datas_index.append(i)
else:
pass
print(thing.loc[datas_index, :])
collect_datas = thing.iloc[datas_index, :3]
collect_datas.hold = collect_datas.hold.apply(lambda x: float(x))
print(collect_datas)
name = collect_datas['name']
collect_datas.drop(labels=['name'], axis=1, inplace=True)
collect_datas.insert(2, 'name', name)
print(collect_datas.get_values())
tr = pyecharts.ThemeRiver(
"TOP 10 hold 河流图 ",
height=800,
width=800,
)
tr.add(name.get_values(),
collect_datas.get_values(),
is_label_show=False)
#trmtimeline.add(tr,"%s"%y,)
page = pyecharts.Page()
# page.add(trmtimeline)
page.add(timeline)
page.add(tr)
'''
them_dates=[]
for i in thing.index:
print (i)
collect_datas=thing[thing.index==i].iloc[:, 0:3]
name=collect_datas['name']
collect_datas.drop(labels=['name'], axis=1,inplace = True)
collect_datas.insert(2, 'name', name)
print (collect_datas.get_values())
'''
print('time line done')
return page
def plot_monitor(dic_psi,
dic_ks,
dic_stats,
cols=None,
path='reportsource/model_monitor.html'):
if not os.path.exists(os.path.dirname(os.path.abspath(path))):
os.makedirs(os.path.dirname(os.path.abspath(path)))
page = pyecharts.Page()
for k in dic_psi.keys():
temp = pd.read_json(json.dumps(dic_psi[k],
cls=MyEncoder)).sort_values('date')
temp = temp[temp.volume > 100]
line = pyecharts.Line(k + ' PSI监控')
line.add('psi',
temp.date,
temp['psi'],
xaxis_rotate=45,
is_datazoom_show=True,
tooltip_tragger='axis',
datazoom_type='both',
legend_pos='center',
legend_orient='horizontal')
line_pass = pyecharts.Line()
line_pass.add('pass_ratio',
temp.date,
temp['pass_ratio'],
xaxis_rotate=45,
is_datazoom_show=True,
tooltip_tragger='axis',
datazoom_type='both',
legend_pos='center',
legend_orient='horizontal')
bar = pyecharts.Bar()
bar.add('volume',
temp.date,
temp['volume'],
xaxis_rotate=45,
is_datazoom_show=True,
tooltip_tragger='axis',
datazoom_type='both',
legend_pos='center',
legend_orient='horizontal')
overlap = pyecharts.Overlap()
overlap.add(line)
overlap.add(line_pass)
overlap.add(bar, yaxis_index=1, is_add_yaxis=True)
page.add(overlap)
for k in dic_ks.keys():
line = pyecharts.Line('%s features ks monitor' % k)
for f in dic_ks[k].keys():
temp = pd.read_json(json.dumps(dic_ks[k][f], cls=MyEncoder))
temp = temp[temp['volume'] > 100]
line.add(f,
temp.date,
temp.ks,
xaxis_rotate=90,
is_datazoom_show=True,
tooltip_tragger='axis',
datazoom_type='both',
legend_pos='right',
legend_orient='vertical')
page.add(line)
for k in dic_stats.keys():
if cols is None:
cols_f = dic_stats[k].keys()
else:
cols_f = [f for f in cols if f in dic_stats[k].keys()]
for f in cols_f:
line = pyecharts.Line('%s features stats monitor: ' % k + f)
temp = pd.DataFrame(dic_stats[k][f]).sort_values('date')
stats_name = temp.iloc[0]['tr_v'].keys()
for n in stats_name:
temp_te_v = temp.te_v.apply(pd.Series)
temp_tr_v = temp.tr_v.apply(pd.Series)
line.add(n,
temp.date,
temp_te_v[n],
is_datazoom_show=True,
tooltip_tragger='axis',
datazoom_type='both',
legend_pos='right',
legend_orient='vertical')
line.add(n + '_benchmark',
temp.date,
temp_tr_v[n],
is_datazoom_show=True,
tooltip_tragger='axis',
datazoom_type='both',
legend_pos='right',
legend_orient='vertical')
page.add(line)
page.render(path)
return None
overlap = echarts.Overlap(width=1200, height=600)
# 默认不新增 x y 轴，并且 x y 轴的索引都为 0
overlap.add(bar)
# 新增一个 y 轴，此时 y 轴的数量为 2，第二个 y 轴的索引为 1（索引从 0 开始），所以设置 yaxis_index = 1
# 由于使用的是同一个 x 轴，所以 x 轴部分不用做出改变
overlap.add(line, yaxis_index=1, is_add_yaxis=True)
overlap.render(path='tmp/overlap.html')
# Page：同一网页按顺序展示多图
if __name__ == '__main__':
import pyecharts as echarts
# 示例
if __name__ == '__main__':
page = echarts.Page() # step 1
# bar
attr = ["衬衫", "羊毛衫", "雪纺衫", "裤子", "高跟鞋", "袜子"]
v1 = [5, 20, 36, 10, 75, 90]
v2 = [10, 25, 8, 60, 20, 80]
bar = echarts.Bar("柱状图数据堆叠示例")
bar.add("商家A", attr, v1, is_stack=True)
bar.add("商家B", attr, v2, is_stack=True)
page.add(bar) # step 2
# scatter3D
import random
data = [
[random.randint(0, 100),
def plot_station_3ntends(re_dic, to_path):
re_dic = add_whereadd_redic(re_dic)
dic_station, dic_smarttime = get_station_detail_(re_dic)
re_dic['stname'] = '%s-%s' % (dic_station['city'], dic_station['stname'])
mapp = {'30':'250', '14':'251', '7':'252', 'dyn':'253'}
re_dic['tar_id'] = mapp[re_dic['N']]
dw = my_(config.MYSQL_BI_RW_ENV)
sql_mer = "SELECT * FROM profile.pr_stations_user_dis2_ where mer_id = %(mer_id)s and tar_id = %(tar_id)s %(whereadd)s" % re_dic
sql_all = "SELECT * FROM profile.pr_platforms_user_dis2_ where mer_id = 0 and tar_id = %(tar_id)s %(whereadd)s" % re_dic
data_mer = dw.to_dataframe(sql_mer)
data_all = dw.to_dataframe(sql_all)
page = pyecharts.Page(page_title='3N分析',)
tem_dic = dict(label_text_size=12, is_label_show=False,
is_fill=True, area_opacity=0.12, line_width=2, is_smooth=True, mark_line=["average"]
,xaxis_interval=1, xaxis_rotate=45)
line1 = pyecharts.Line("当日加油的隔(N-2N)天回流占比:", "%(stname)s(%(mer_id)s)-(N=%(N)s)" % re_dic, width=1500, height=600)
X = data_all.query('spr_id==11')['batch_date'].tolist()
N1today_mer_rate = np.array(data_mer.query('spr_id==9')['value_c']) / np.array(data_mer.query('spr_id==8')['value_c'])
N1today_all_rate = np.array(data_all.query('spr_id==9')['value_c']) / np.array(data_all.query('spr_id==8')['value_c'])
Y_mer = N1today_mer_rate.round(4)
Y_all = N1today_all_rate.round(4)
conb_ = np.append(Y_mer,Y_all)
tem_dic.update({'yaxis_min':conb_.min()*0.6})
lineadd_sted(line1, dic_smarttime, X, conb_.max() * 1.2)
line1.add("油站", X, Y_mer, **tem_dic)
line1.add("全平台", X, Y_all, **tem_dic)
page.add(line1)
line2 = pyecharts.Line("当日加油的隔(2N-3N)天回流占比:", "%(stname)s(%(mer_id)s)-(N=%(N)s)" % re_dic, width=1500, height=600)
X = data_all.query('spr_id==11')['batch_date']
N2today_mer_rate = np.array(data_mer.query('spr_id==10')['value_c']) / np.array(data_mer.query('spr_id==8')['value_c'])
N2today_all_rate = np.array(data_all.query('spr_id==10')['value_c']) / np.array(data_all.query('spr_id==8')['value_c'])
Y_mer = N2today_mer_rate.round(4)
Y_all = N2today_all_rate.round(4)
conb_ = np.append(Y_mer, Y_all)
tem_dic.update({'yaxis_min': conb_.min()*0.6})
lineadd_sted(line2, dic_smarttime, X, conb_.max() * 1.2)
line2.add("油站:", X, Y_mer, **tem_dic)
line2.add("全平台", X, Y_all, **tem_dic)
page.add(line2)
line2 = pyecharts.Line("最近周期加油用户中睡眠回流占比:", "%(stname)s(%(mer_id)s)-(N=%(N)s)" % re_dic, width=1500, height=600)
N3_mer_rate = np.array(data_mer.query('spr_id==4')['value_c']) / np.array(data_mer.query('spr_id==2')['value_c'])
N3_all_rate = np.array(data_all.query('spr_id==4')['value_c']) / np.array(data_all.query('spr_id==2')['value_c'])
Y_mer = N3_mer_rate.round(4)
Y_all = N3_all_rate.round(4)
conb_ = np.append(Y_mer, Y_all)
tem_dic.update({'yaxis_min': conb_.min() * 0.6})
lineadd_sted(line2, dic_smarttime, X, conb_.max() * 1.2)
tem_dic.update({'yaxis_min': conb_.min()*0.6})
line2.add("油站", X, Y_mer, **tem_dic)
line2.add("全平台", X, Y_all, **tem_dic)
page.add(line2)
line2 = pyecharts.Line("睡眠回流占比:", "%(stname)s(%(mer_id)s)-(N=%(N)s)" % re_dic, width=1500, height=600)
N3_mer_rate = np.array(data_mer.query('spr_id==4')['value_c']) / np.array(data_mer.query('spr_id==3')['value_c'])
N3_all_rate = np.array(data_all.query('spr_id==4')['value_c']) / np.array(data_all.query('spr_id==3')['value_c'])
Y_mer = N3_mer_rate.round(4)
Y_all = N3_all_rate.round(4)
conb_ = np.append(Y_mer, Y_all)
tem_dic.update({'yaxis_min': conb_.min() * 0.6})
lineadd_sted(line2, dic_smarttime, X, conb_.max() * 1.2)
tem_dic.update({'yaxis_min': conb_.min() - 0.04})
line2.add("油站", X, Y_mer, **tem_dic)
line2.add("全平台", X, Y_all, **tem_dic)
page.add(line2)
line2 = pyecharts.Line("有睡眠趋势的用户占比:", "%(stname)s(%(mer_id)s)-(N=%(N)s)" % re_dic, width=1500, height=600)
N3_mer_rate = np.array(data_mer.query('spr_id==5')['value_c']) / np.array(data_mer.query('spr_id==1')['value_c'])
N3_all_rate = np.array(data_all.query('spr_id==5')['value_c']) / np.array(data_all.query('spr_id==1')['value_c'])
Y_mer = N3_mer_rate.round(4)
Y_all = N3_all_rate.round(4)
conb_ = np.append(Y_mer, Y_all)
tem_dic.update({'yaxis_min': conb_.min() * 0.6})
lineadd_sted(line2, dic_smarttime, X, conb_.max() * 1.2)
tem_dic.update({'yaxis_min': conb_.min() - 0.04})
line2.add("油站", X, Y_mer, **tem_dic)
line2.add("全平台", X, Y_all, **tem_dic)
page.add(line2)
page.render(to_path)
def parser_job(i):
# 工作的链接
job_url = re.search('href="(.*?)"', i).group(1)
name = re.search("<h3>(.*?)</h3>", i).group(1)
add = re.search("<em>(.*?)</em>", i).group(1)
money = re.search('"money">(.*?)</span>', i).group(1)
jy, xl = re.search('-->([\s\S]*?)\n', i).group(1).split(" / ")
return {"name":name,"add":add,"job_url":job_url,"money":money,"jy":jy,"xl":xl}
# 指定需要查看的技术方向
jobs = ["Java","Python"]
# 创建一个页面
page = pyecharts.Page()
for url in urls:
if url[1] not in jobs:
continue
# 存储解析完成的字典
datas = []
for j in range(1,6):
res = requests.get(url[0] + str(j))
data = re.findall("""<a class="position_link"([\s\S]*?)<div class="company">""",res.text)
for i in data:
dic = parser_job(i)
if "{" in dic["xl"]:
continue
datas.append(dic)