def test_boxplot_item_base(fake_writer):
x_axis = ["expr1", "expr2"]
v1 = [
[850, 740, 900, 1070, 930, 850, 950, 980, 980, 880, 1000, 980],
[960, 940, 960, 940, 880, 800, 850, 880, 900, 840, 830, 790],
]
v2 = [
[890, 810, 810, 820, 800, 770, 760, 740, 750, 760, 910, 920],
[890, 840, 780, 810, 760, 810, 790, 810, 820, 850, 870, 870],
]
c = Boxplot()
series_a = [
opts.BoxplotItem(name=x_axis[0], value=d) for d in c.prepare_data(v1)
]
series_b = [
opts.BoxplotItem(name=x_axis[1], value=d) for d in c.prepare_data(v2)
]
c.add_xaxis(xaxis_data=x_axis).add_yaxis("A", series_a).add_yaxis(
"B", series_b)
c.render()
_, content = fake_writer.call_args[0]
assert_equal(c.theme, "white")
assert_equal(c.renderer, "canvas")
def boxplot(self, column:str):
if self.check_col_type(column, "number"):
series = self.data[column]
boxplot = Boxplot()
boxplot.add_xaxis([column]).add_yaxis("", boxplot.prepare_data([list(series)]))
boxplot.set_global_opts(title_opts=opts.TitleOpts(title="{}-箱线图".format(column)))
return boxplot.dump_options()
else:
return None
def boxpolt_base() -> Boxplot:
v_sophomore = [[1.1, 2.2, 2.6, 3.2, 3.7, 4.2, 4.7, 4.7, 5.5, 6.3, 8.0],
[2.5, 2.5, 2.8, 3.2, 3.7, 4.2, 4.7, 4.7, 5.5, 6.3, 7.0]]
v_junior = [[3.6, 3.7, 4.7, 4.9, 5.1, 5.2, 5.3, 5.4, 5.7, 5.8, 5.8],
[3.6, 3.7, 4.7, 4.9, 5.1, 5.2, 5.3, 5.4, 5.7, 5.8, 5.8]]
# 最小值，下四分位数，中位数、上四分位数、最大值
# [min, Q1, median (or Q2), Q3, max]
c = (
Boxplot().add_xaxis(['寒假作业', '暑假作业']).add_yaxis(
'大二队员', Boxplot.prepare_data(v_sophomore)).add_yaxis(
'大三队员', Boxplot.prepare_data(v_junior)).set_series_opts(
label_opts=opts.LabelOpts(is_show=True)).set_global_opts(
title_opts=opts.TitleOpts(title='ACM集训队祖传练习完成时长离散度'),
xaxis_opts=opts.AxisOpts(name='单位：小时'),
legend_opts=opts.LegendOpts(
is_show=True)).reversal_axis() #翻转XY轴
)
return c
def boxpolt_base() -> Boxplot:
v1 = [
[850, 740, 900, 1070, 930, 850, 950, 980, 980, 880] +
[1000, 980, 930, 650, 760, 810, 1000, 1000, 960, 960],
[960, 940, 960, 940, 880, 800, 850, 880, 900] +
[840, 830, 790, 810, 880, 880, 830, 800, 790, 760, 800],
]
v2 = [
[890, 810, 810, 820, 800, 770, 760, 740, 750, 760] +
[910, 920, 890, 860, 880, 720, 840, 850, 850, 780],
[890, 840, 780, 810, 760, 810, 790, 810, 820, 850, 870] +
[870, 810, 740, 810, 940, 950, 800, 810, 870],
]
c = Boxplot()
c.add_xaxis(["expr1", "expr2"])\
.add_yaxis("A", c.prepare_data(v1))\
.add_yaxis("B", c.prepare_data(v2)
).set_global_opts(title_opts=opts.TitleOpts(title="BoxPlot-基本示例"))
return c
def boxplot_product_price_base():
"""
此函数用于获取产品价格的箱线图的参数。
Returns
-------
c : TYPE-echarts parameters
return echarts parameters.
"""
# data query
dataX, dataY = product_price_query()
# Declare objects, render pictures
c = Boxplot()
c.add_xaxis(dataX)
c.add_yaxis("Pricing",c.prepare_data(dataY))
c.set_global_opts(title_opts=opts.TitleOpts(title="Distribution of Product Pricing", subtitle="Unit:$"))
return c
def test_boxpolt_base():
v1 = [
[850, 740, 900, 1070, 930, 850, 950, 980, 980, 880, 1000, 980],
[960, 940, 960, 940, 880, 800, 850, 880, 900, 840, 830, 790],
]
v2 = [
[890, 810, 810, 820, 800, 770, 760, 740, 750, 760, 910, 920],
[890, 840, 780, 810, 760, 810, 790, 810, 820, 850, 870, 870],
]
c = Boxplot()
c.add_xaxis(["expr1", "expr2"]).add_yaxis("A", c.prepare_data(v1)).add_yaxis(
"B", c.prepare_data(v2)
)
assert c.theme == "white"
assert c.renderer == "canvas"
c.render("render.html")
def test_boxpolt_base(fake_writer):
v1 = [
[850, 740, 900, 1070, 930, 850, 950, 980, 980, 880, 1000, 980],
[960, 940, 960, 940, 880, 800, 850, 880, 900, 840, 830, 790],
]
v2 = [
[890, 810, 810, 820, 800, 770, 760, 740, 750, 760, 910, 920],
[890, 840, 780, 810, 760, 810, 790, 810, 820, 850, 870, 870],
]
c = Boxplot()
c.add_xaxis(["expr1", "expr2"]).add_yaxis("A",
c.prepare_data(v1)).add_yaxis(
"B", c.prepare_data(v2))
c.render()
_, content = fake_writer.call_args[0]
eq_(c.theme, "white")
eq_(c.renderer, "canvas")
def box_plot(data: dict,
title: str = "箱线图",
width: str = "900px",
height: str = "680px") -> Boxplot:
"""
:param data: 数据
样例：
data = {
"expr 0": [960, 850, 830, 880],
"expr 1": [960, 850, 830, 880],
}
:param title:
:param width:
:param height:
:return:
"""
x_data = []
y_data = []
for k, v in data.items():
x_data.append(k)
y_data.append(v)
init_opts = opts.InitOpts(page_title=title, width=width, height=height)
chart = Boxplot(init_opts=init_opts)
chart.add_xaxis(xaxis_data=x_data)
chart.add_yaxis(series_name="", y_axis=y_data)
chart.set_global_opts(title_opts=opts.TitleOpts(pos_left="center", title=title),
tooltip_opts=opts.TooltipOpts(trigger="item", axis_pointer_type="shadow"),
xaxis_opts=opts.AxisOpts(
type_="category",
boundary_gap=True,
splitarea_opts=opts.SplitAreaOpts(is_show=False),
axislabel_opts=opts.LabelOpts(formatter="{value}"),
splitline_opts=opts.SplitLineOpts(is_show=False),
),
yaxis_opts=opts.AxisOpts(
type_="value",
name="",
splitarea_opts=opts.SplitAreaOpts(
is_show=True, areastyle_opts=opts.AreaStyleOpts(opacity=1)
)
))
return chart
def draw_project_cost_boxplot():
row_dict = request_data.get_project_cost()
return_code = row_dict['code']
if return_code == 0:
x_data = ['Success', 'Failed']
success_data = row_dict['stateProjectCost']['state']['success']
y_data1 = [[success_data['min'], success_data['25%'], success_data['mean'], success_data['75%'],
success_data['max']]]
failed_data = row_dict['stateProjectCost']['state']['failed']
y_data2 = [[failed_data['min'], failed_data['25%'], failed_data['mean'], failed_data['75%'], failed_data['max']]]
print('boxplot')
print(x_data)
print(y_data1)
print(y_data2)
project_cost_boxplot = Boxplot()
project_cost_boxplot.add_xaxis(x_data)
project_cost_boxplot.add_yaxis("成功项目", y_data1)
project_cost_boxplot.add_yaxis("失败项目", y_data2)
project_cost_boxplot.set_global_opts(title_opts=opts.TitleOpts(title="项目花费"))
return project_cost_boxplot
else:
return 0
def find_distribution_render(x,Y,x_name,query,table_path,answer):
Max=0
y=[]
data=[]
label=[]
boxplot = Boxplot()
boxplot.add_xaxis(xaxis_data=label)
for i in range(len(Y)):
data.append(Y[i][1])
label.append(Y[i][0])
# boxplot.add_yaxis(series_name=Y[i][0],y_axis=boxplot.prepare_data(data))
for j in Y[i][1]:
y.append(j)
boxplot.add_yaxis(series_name="", y_axis=boxplot.prepare_data(data))
boxplot.set_global_opts(
yaxis_opts=opts.AxisOpts(name="Distribution",axislabel_opts=opts.LabelOpts(font_size="100%"),name_textstyle_opts=opts.TextStyleOpts(font_size="100%")),
legend_opts=opts.LegendOpts(is_show=False),
xaxis_opts=opts.AxisOpts(axislabel_opts=opts.LabelOpts(rotate=40,interval=0,font_size='100%'),name="Category",name_textstyle_opts=opts.TextStyleOpts(font_size="100%")),
graphic_opts=[opts.GraphicText(
graphic_item=opts.GraphicItem(
left="center",
top="bottom",
z=100,
),
graphic_textstyle_opts=opts.GraphicTextStyleOpts(
# 可以通过jsCode添加js代码，也可以直接用字符串
text=['\n' + "Q:" + ' ' + query + '\n'+"\n" + 'A:' + ' ' + answer],
font="14px Microsoft YaHei",
graphic_basicstyle_opts=opts.GraphicBasicStyleOpts(
fill="#333"
)
)
)]
)
bar=Bar({"theme": ThemeType.MACARONS})
bar.add_xaxis(xaxis_data=x)
for i in range(len(Y)):
max_index, max_number = max(enumerate(Y[i][1]), key=operator.itemgetter(1))
min_index, min_number = min(enumerate(Y[i][1]), key=operator.itemgetter(1))
if max_number>Max:
Max=max_number
bar.add_yaxis(Y[i][0],label_opts=opts.LabelOpts(is_show=False),y_axis=Y[i][1],markpoint_opts=opts.MarkPointOpts(
data=[opts.MarkPointItem(coord=[max_index,max_number*1.01],value="MAX",name="最大值",itemstyle_opts=opts.ItemStyleOpts(opacity=0.6)),opts.MarkPointItem(coord=[min_index,min_number*1.05],value="MIN",name="最小值",itemstyle_opts=opts.ItemStyleOpts(opacity=0.6))]
),
markline_opts=opts.MarkLineOpts(
data=[opts.MarkLineItem(type_="average",name="平均值")]
))
bar.set_global_opts(
datazoom_opts=[opts.DataZoomOpts(range_start=10, range_end=90),
opts.DataZoomOpts(type_="inside")],
yaxis_opts=opts.AxisOpts( axislabel_opts=opts.LabelOpts(font_size="100%"),
name_textstyle_opts=opts.TextStyleOpts(font_size="100%")),
xaxis_opts=opts.AxisOpts(axislabel_opts=opts.LabelOpts(rotate=40, interval=0, font_size='100%'),
name=x_name, name_textstyle_opts=opts.TextStyleOpts(font_size="100%")),
graphic_opts=[opts.GraphicText(
graphic_item=opts.GraphicItem(
left="center",
top="bottom",
),
graphic_textstyle_opts=opts.GraphicTextStyleOpts(
# 可以通过jsCode添加js代码，也可以直接用字符串
text=['\n' + "Q:" + ' ' + query + '\n' + "\n"+'A:' + ' ' + answer],
font="14px Microsoft YaHei",
graphic_basicstyle_opts=opts.GraphicBasicStyleOpts(
fill="#333"
)
)
)]
)
grid = Grid( init_opts=opts.InitOpts(
width="100%",
height="100%",
renderer= globals.RenderType.SVG,))
grid.add(boxplot, grid_opts={'left':'20%','bottom':'34%'})
grid1 = Grid( init_opts=opts.InitOpts(
width="100%",
height="100%",
renderer= globals.RenderType.SVG,))
grid1.add(bar,grid_opts={'left':'20%','bottom':'34%'})
option1=grid.dump_options_with_quotes()
option1=json.loads(option1)
option2=grid1.dump_options_with_quotes()
option2=json.loads(option2)
option={"option":[option1,option2],"query":query}
return option
y1 = [83, 88, 93, 79, 86, 89, 95, 90, 94, 92]
y2 = [77, 79, 74, 82, 73, 78, 75, 85, 83, 81]
y3 = [94, 93, 97, 89, 92, 97, 91, 85, 82, 96]
y4 = [82, 81, 79, 85, 88, 73, 78, 82, 75, 79]
y5 = [71, 78, 74, 73, 79, 77, 75, 74, 75, 81]
y6 = [62, 65, 72, 66, 78, 73, 66, 68, 61, 63]
# 绘制箱线图
# 注意用yi生成各科的[min, Q2, mid, Q1, max]数据，用于后续绘制箱线图
# 绘图的其它需求
# 1. 主题为马卡龙画风(macarons)
# 2. 让箱线图的须更长，数据分布更易于观察：把y轴的最小值设为数据的最小值，而不是从0开始
# （这一步怎么设置的，建议查一下文档，写得明明白白）
# 3. 全图标题“各科成绩”
boxplot = (
Boxplot(init_opts=opts.InitOpts(theme="macarons")).set_global_opts(
title_opts=opts.TitleOpts(title="各科成绩"),
yaxis_opts=opts.AxisOpts(min_="dataMin") # dataMin怎么来的，官方文档上去查，写得明明白白
).add_xaxis(xaxis_data=xData))
# 添加y轴数据
boxplot.add_yaxis(series_name="",
y_axis=boxplot.prepare_data([y1, y2, y3, y4, y5, y6]))
# 保存路径
resultPath = "./result"
if not os.path.exists(path=resultPath):
os.mkdir(path=resultPath)
# 保存文件名
resultFileName = "scores_boxplot.html"
# 执行保存
boxplot.render(path=os.path.join(resultPath, resultFileName))
dom1, dom2, dom3 = [], [], []
list1 = []
for date, AQI in zip(data['Date'], data['AQI']):
year = date.split('-')[0]
if year in ['2015', '2016']:
dom1.append(AQI)
elif year in ['2017', '2018']:
dom2.append(AQI)
elif year in ['2019', '2020']:
dom3.append(AQI)
x = ['2015-2016年', '2017-2018年', '2019-2020年']
y = [dom1, dom2, dom3]
boxplot = Boxplot()
boxplot.add_xaxis(x)
boxplot.add_yaxis('Delhi', boxplot.prepare_data(y))
boxplot.set_global_opts(title_opts=opts.TitleOpts(
title='2015-2020印度德里AQI箱型图',
pos_left='center',
),
legend_opts=opts.LegendOpts(
pos_left='left',
orient='vertical',
))
boxplot.render('boxplot_AQI.html')
def paintBoxPlot(datadict, xtitle, ytitle, title):
graph = Boxplot(init_opts=opts.InitOpts(page_title=title))
graph.set_global_opts(
title_opts=opts.TitleOpts(title=title),
yaxis_opts=opts.AxisOpts(name='Accuracy')
)
graph.add_xaxis([xtitle])
for key, value in sorted(datadict.items()):
# 为了画图的标准性，小于5人的年级图片将不会被加入盒图
if len(value) > 4:
graph.add_yaxis(key, graph.prepare_data([value]))
graph.render('buctoj2_boxplot.html')
# In[94]:
look_support_money = article[['看一看', '点赞', '赞赏']]
# 箱型图绘制
from pyecharts import options as opts
from pyecharts.charts import Boxplot
v1 = [
list(look_support_money['看一看']),
list(look_support_money['点赞']),
list(look_support_money['赞赏'])
]
c = Boxplot(init_opts=opts.InitOpts(theme=ThemeType.VINTAGE, chart_id=4))
c.add_xaxis(["看一看", "点赞", '赞赏'])
c.add_yaxis("", c.prepare_data(v1))
c.set_global_opts(title_opts=opts.TitleOpts(title="看一看,点赞,赞赏分布"))
c.render("../output/看一看,点赞,赞赏.html")
c.render_notebook()
# ## 文章类型占比
# In[95]:
kind = article['文章类型'].value_counts()
# 绘制玫瑰图
from pyecharts import options as opts
from pyecharts.charts import Pie
# In[138]:
# vis
from pyecharts import options as opts
from pyecharts.charts import Boxplot
v1 = [
[850, 740, 900, 1070, 930, 850, 950, 980, 980, 880, 1000, 980],
[960, 940, 960, 940, 880, 800, 850, 880, 900, 840, 830, 790],
]
v2 = [
[890, 810, 810, 820, 800, 770, 760, 740, 750, 760, 910, 920],
[890, 840, 780, 810, 760, 810, 790, 810, 820, 850, 870, 870],
]
c = Boxplot()
c.add_xaxis(["expr1", "expr2"])
c.add_yaxis("A", c.prepare_data(v1))
c.add_yaxis("B", c.prepare_data(v2))
c.set_global_opts(title_opts=opts.TitleOpts(title="BoxPlot-基本示例"))
c.render_notebook()
# ## WordCloud
# In[156]:
# vis
import pyecharts.options as opts
from pyecharts.charts import WordCloud
import jieba.analyse
import os from pyecharts.charts import Boxplot, Grid import pyecharts.options as opts # 准备原始数据（为方便直观表达绘制过程，这里读表的过程先跳过） # x轴数据 xData = ["15℃", "20℃", "25℃", "30℃", "35℃"] # 用于绘制y轴的原始数据（假设一组yi对应一个x_data下的实验组测得的一组数据） y1 = [383, 312, 348, 388, 375, 389, 432, 373, 377, 384] y2 = [412, 432, 421, 354, 378, 392, 407, 413, 415, 405] y3 = [400, 388, 397, 352, 423, 447, 378, 377, 395, 387] y4 = [278, 288, 254, 213, 256, 278, 267, 256, 278, 266] y5 = [213, 215, 234, 232, 278, 188, 223, 234, 225, 231] # 绘制箱线图 boxplot = Boxplot() # 生成绘箱线图图用的y轴数据（即每一组yi）的[下限值，下四分位点，中位数，上四分位点，上限值] yData = boxplot.prepare_data([y1, y2, y3, y4, y5]) # print(y_data) # 得到绘制5组温度下对应箱线图的[下限值，下四分位点，中位数，上四分位点，上限值]数据 # [[312, 366.75, 380.0, 388.25, 432], [354, 388.5, 409.5, 416.5, 432], [352, 377.75, 391.5, 405.75, 447], # [213, 255.5, 266.5, 278.0, 288], [188, 214.5, 228.0, 234.0, 278]] # 执行绘图 # 添加x轴数据 boxplot.add_xaxis(xaxis_data=xData) # 添加y轴数据 boxplot.add_yaxis(series_name="实验组数据", y_axis=yData) # 保存路径 resultPath = "./result"
[
880, 880, 880, 860, 720, 720, 620, 860, 970, 950, 880, 910, 850,
870, 840, 840, 850, 840, 840, 840
],
[
890, 810, 810, 820, 800, 770, 760, 740, 750, 760, 910, 920, 890,
860, 880, 720, 840, 850, 850, 780
],
[
890, 840, 780, 810, 760, 810, 790, 810, 820, 850, 870, 870, 810,
740, 810, 940, 950, 800, 810, 870
]]
scatter_data = [650, 620, 720, 720, 950, 970]
box_plot = Boxplot()
box_plot = (box_plot.add_xaxis(
xaxis_data=["expr 0", "expr 1", "expr 2", "expr 3", "expr 4"]).add_yaxis(
series_name="", y_axis=box_plot.prepare_data(y_data)).set_global_opts(
title_opts=opts.TitleOpts(pos_left="center",
title="Michelson-Morley Experiment"),
tooltip_opts=opts.TooltipOpts(trigger="item",
axis_pointer_type="shadow"),
xaxis_opts=opts.AxisOpts(
type_="category",
boundary_gap=True,
splitarea_opts=opts.SplitAreaOpts(is_show=False),
axislabel_opts=opts.LabelOpts(formatter="expr {value}"),
splitline_opts=opts.SplitLineOpts(is_show=False),
),
yaxis_opts=opts.AxisOpts(
def salary(self):
sql = 'SELECT workYear,replace(salary,\'k\',\'\') s FROM jobs group by workYear,salary order by workYear'
results = self.query(sql)
sum = {}
for r in results:
rs = r[1].split('-')
a = sum.get(r[0], [])
a.extend(rs)
sum[r[0]] = a
for k in sum:
numbers = list(map(int, sum[k]))
v = list(set(numbers))
sum[k] = v
print(list(sum.values()))
c = Boxplot()
c.add_xaxis(list(sum.keys()))
c.add_yaxis("薪资与工作经验", c.prepare_data(list(sum.values())))
c.set_global_opts(title_opts=opts.TitleOpts(title="薪资与工作经验"))
c.render("拉勾薪资.html")
plotData = {"一等舱": [], "二等舱": [], "三等舱": []}
# 扫描原始数据
for pClass, age in orgData:
if pClass == 1:
plotData["一等舱"].append(age)
elif pClass == 2:
plotData["二等舱"].append(age)
elif pClass == 3:
plotData["三等舱"].append(age)
# 绘制船舱等级与年龄关系的箱线图，需求：
# 1. 主题为light
# 2. 标题为“舱位与年龄的分布”
# 3. y轴起点是上述y轴所有数据的最小值
boxplot = (Boxplot(init_opts=opts.InitOpts(theme="light")).set_global_opts(
title_opts=opts.TitleOpts(title="舱位与年龄的分布"),
yaxis_opts=opts.AxisOpts(min_="dataMin")).add_xaxis(
xaxis_data=list(plotData.keys())))
# 获取箱线图y轴[min, Q2, mid, Q1, max]序列
yData = boxplot.prepare_data(
items=[plot_value for plot_value in plotData.values()])
print(yData)
boxplot.add_yaxis(series_name="", y_axis=yData)
# 保存路径
resultPath = "./result"
if not os.path.exists(path=resultPath):
os.mkdir(path=resultPath)
# 保存文件名
resultFileName = "titanic_boxplot.html"
# 执行保存
boxplot.render(path=os.path.join(resultPath, resultFileName))
def pollsummary(self,subset='masterCandle',dat_cursor=None,renderfile=False):
# 初始化交易时间x轴
heartbeat = 300
# 如果未提供复盘时间游标，则默认使用最近交易日至收盘收盘时间的数据
if dat_cursor is None:
dat_cursor = datetime.now().astimezone(timezone(cfg.STR_TIMEZONE))
int_latestViewDate = self.da.int_findClosestTradeDate(datetime.strftime(dat_cursor,'%Y%m%d') )
int_cursorDateTime = int(str(int_latestViewDate)+'150100')
int_cursorDateStart = int(str(int_latestViewDate)+'000000')
# 如提供复盘游标时间，则该时间为游标卡尺的最后时间，如日期为非交易日则自动调整为最近交易日的收盘时间
else:
int_cursorDate = datetime.strftime(dat_cursor,'%Y%m%d')
int_latestViewDate = self.da.int_findClosestTradeDate(int_cursorDate)
if int_cursorDate==int_latestViewDate: # 如相等说明提供日期为有效交易日，则使用提供的小时分钟秒为游标终点时间，否则将自动替换为最近交易日的收盘时间点
int_cursorDateTime = int(datetime.strftime(dat_cursor,'%Y%m%d%H%M%S'))
int_cursorDateStart = int(datetime.strftime(dat_cursor,'%Y%m%d')+'000000')
else:
int_cursorDateTime = int(str(int_latestViewDate) +'150100')
int_cursorDateStart = int(str(int_latestViewDate) +'000000')
beatRange = Radar.pop_pulseRange(int_latestViewDate,heartbeat)
df_stockPoll = pd.read_sql(DataAgent.dbsession.query(Stock_poll).filter(Stock_poll.time_index>=int_cursorDateStart,
Stock_poll.time_index<=int_cursorDateTime,
Stock_poll.volume!=0).statement,
DataAgent.dbsession.bind)
df_indexPoll = pd.read_sql(DataAgent.dbsession.query(Index_poll).filter(Index_poll.time_index>=int_cursorDateStart,
Index_poll.time_index<=int_cursorDateTime,
Index_poll.volume!=0).statement,
DataAgent.dbsession.bind)
'''
data_store = libs.tryOpenH5(cfg.H5_FILE_POLL,mode='r')
if 'stockPoll' in data_store:
df_stockPoll = data_store['stockPoll']
df_stockPoll = df_stockPoll[df_stockPoll['volume']!=0] # 过滤掉数据中非交易中的证券
df_stockPoll = df_stockPoll[(df_stockPoll['time_index']<=int_cursorDateTime)&(df_stockPoll['time_index']>=int_cursorDateStart)] # 过滤掉和画图期间无关的数据
else:
df_stockPoll = pd.DataFrame()
df_indexPoll = data_store['indexPoll'] if 'indexPoll' in data_store else pd.DataFrame()
data_store.close()
'''
data_store = libs.tryOpenH5(cfg.H5_FILE_PRE,mode='r')
# 如果按照输入日期没有找到当日任何盘中数据，则从入库历史数据中找到当日收盘数据用于填充
if len(df_indexPoll)==0:
df_indexPoll = data_store['masterCandle'][data_store['masterCandle']['trade_date']==int_latestViewDate]
df_indexPoll['time_index'] = int_cursorDateTime
df_indexPoll['pct_change'] = round(df_indexPoll['adj_close']/df_indexPoll['pre_close']*100,2)
df_indexPoll['per'] = np.nan
if len(df_stockPoll)==0:
df_stockPoll = data_store['masterCandle'][data_store['masterCandle']['trade_date']==int_latestViewDate]
df_stockPoll['time_index'] = int_cursorDateTime
df_stockPoll['pct_change'] = round((df_stockPoll['adj_close']-df_stockPoll['pre_close'])/df_stockPoll['pre_close']*100,2)
df_stockPoll.rename(columns={'pe': 'per'}, inplace=True)
# 如果入库历史数据中仍未找到任何数据则退出分析程序
if any([len(df_stockPoll)==0,len(df_indexPoll)==0]):
data_store.close()
print('stockPoll or indexPoll does not exist for the cursor date in the offline data store, process exiting...')
return False
''' ------读取security code级别的聚合数据------- '''
if subset=='masterCandle':
pass #data_store = libs.tryOpenH5(cfg.H5_FILE_PRE,mode='r')
else:
data_store.close() # 关闭之前打开的master data file
data_store = libs.tryOpenH5('{}{}.dat'.format(cfg.H5_FILE_PATH,subset),mode='r')
df_histRec = data_store['AggByCode']
data_store.close()
self.poll_aggHistRec = df_histRec # sec code 级别的聚合数据，集合包含证券及指数,输出至对象变量
''' ------读取security code级别的聚合结束------- '''
sr_intBeatIndex = pd.Series(beatRange.strftime('%Y%m%d%H%M%S').astype('int64')).rename(index='time_index')
df_index = pd.merge(sr_intBeatIndex,df_indexPoll,how='left',left_on='time_index',right_on='time_index',suffixes=['','1'])
df_stock = pd.merge(sr_intBeatIndex,df_stockPoll,how='left',left_on='time_index',right_on='time_index',suffixes=['','1'])
df_stock = pd.merge(df_stock,df_histRec,how='left',left_on='ts_code',right_index=True,suffixes=['','1'])
self.poll_index = df_index #将结果输出至analyitics对象用于方法外的访问
self.poll_stock = df_stock #将结果输出至analyitics对象用于方法外的访问
# libs.df_csv(cfg.PATH_BUGFILE,(df_stock,))
''' --------所有by time index级别的聚合在这里完成 -----'''
def popPctList(x):
# 移除涨跌幅超过+- 10%的, 新股不受10%涨跌幅限制因此会干扰box chart显示
return [0]*3 if len(x)<3 else [i for i in x if abs(i)<11]
def bigBox(x, up=True):
# 中阳（阴）以上k线数量
if up:
return list(x.loc[(x['box_size']>0) & (x['klineSML'].str.match(DataAgent.re_bigUpBoxPat))]['ts_code'])
else:
return list(x.loc[(x['box_size']<0) & (x['klineSML'].str.match(DataAgent.re_bigDnBoxPat))]['ts_code'])
def breakThrough(x,up=True):
if up: # 找到上穿有效末跌高点的
return list(x.loc[x['close']>=x['validPiv_dnHigh']]['ts_code'])
else: # 找到跌破有效末升低点的
return list(x.loc[x['close']<=x['validPiv_upLow']]['ts_code'])
df_stockBeatGrp = df_stock.groupby('time_index')
df_stockBeatRollup = df_stockBeatGrp.agg({'ts_code': ['count'],
'pct_change': [popPctList],
'per': ['median']}) # function放入中括号内会生成两个层级列名，方便后面join成新列名.如无中括号将只产生一层列名
df_stockBeatRollup.columns = ['_'.join(x) for x in df_stockBeatRollup.columns.ravel()]
df_stockBeatRollup['upThruList'] = df_stockBeatGrp.apply(lambda x: breakThrough(x,up=True)) if len(df_histRec) > 0 else np.nan
df_stockBeatRollup['dnThruList']= df_stockBeatGrp.apply(lambda x: breakThrough(x,up=False)) if len(df_histRec) > 0 else np.nan
df_stockBeatRollup['upThruCount'] = df_stockBeatRollup.apply(lambda row: len(row['upThruList']),axis=1) if len(df_histRec) > 0 else np.nan
df_stockBeatRollup['dnThruCount']= df_stockBeatRollup.apply(lambda row: len(row['dnThruList']),axis=1) if len(df_histRec) > 0 else np.nan
df_stockBeatRollup['upBigBoxList'] = df_stockBeatGrp.apply(lambda x: bigBox(x,up=True))
df_stockBeatRollup['dnBigBoxList'] = df_stockBeatGrp.apply(lambda x: bigBox(x,up=False))
df_stockBeatRollup['upBigBoxCount'] = df_stockBeatRollup.apply(lambda row: len(row['upBigBoxList']),axis=1) if len(df_histRec) > 0 else np.nan
df_stockBeatRollup['dnBigBoxCount'] = df_stockBeatRollup.apply(lambda row: len(row['dnBigBoxList']),axis=1) if len(df_histRec) > 0 else np.nan
''' --------by time index级别的聚合结束 -------------------------------'''
df_stockBeatRollup.astype(dtype= {'ts_code_count':'int32','upThruCount':'int32','dnThruCount':'int32','upBigBoxCount':'int32','dnBigBoxCount':'int32'})
self.poll_aggStkPoll = df_stockBeatRollup.loc[df_stockBeatRollup['ts_code_count']>0] # 只将有效结果输出到analytics对象属性中用于function以外的调用
''' ------------------作图区域---------------------------'''
per_median = round(self.poll_aggStkPoll.loc[self.poll_aggStkPoll.index==self.poll_aggStkPoll.index.max()].per_median.values[0],2)
lst_xaxis = ['{}:{}:{}'.format(i.hour,i.minute,i.second) for i in beatRange]
marketSummaryBoxPlot = Boxplot()
lst_yaxis = np.around(marketSummaryBoxPlot.prepare_data(df_stockBeatRollup['pct_change_popPctList'].values), decimals=2).tolist()
marketSummaryBoxPlot.add_xaxis(lst_xaxis)
marketSummaryBoxPlot.add_yaxis("%", lst_yaxis)
marketSummaryBoxPlot.set_global_opts(title_opts=opts.TitleOpts(title='整体涨跌幅分布 - {} (最后市盈率中位数:{})'.format(int_latestViewDate,str(per_median))),
yaxis_opts=opts.AxisOpts(name="涨幅",min_=-10,max_=10),
xaxis_opts=opts.AxisOpts(name="",
axislabel_opts=opts.LabelOpts(is_show=False),
axisline_opts=opts.AxisLineOpts(is_show=True),
axistick_opts=opts.AxisTickOpts(is_show=True),
splitline_opts=opts.SplitLineOpts(is_show=False),),
legend_opts=opts.LegendOpts(is_show=False),
)
lst_upThruBar = df_stockBeatRollup['upThruCount'].values.tolist()
lst_dnThruBar = df_stockBeatRollup['dnThruCount'].values.tolist()
lst_bigUpBoxBar = df_stockBeatRollup['upBigBoxCount'].values.tolist()
lst_bigDnBoxBar = df_stockBeatRollup['dnBigBoxCount'].values.tolist()
pivotCountChart = (
Bar()
.add_xaxis(lst_xaxis)
.add_yaxis(
series_name="向上突破数量",
yaxis_data=lst_upThruBar,
label_opts=opts.LabelOpts(is_show=False),
itemstyle_opts=opts.ItemStyleOpts(color="#FD625E"),
)
.add_yaxis(
series_name="向下突破数量",
xaxis_index=1,
yaxis_index=1,
yaxis_data=lst_dnThruBar,
label_opts=opts.LabelOpts(is_show=False),
itemstyle_opts=opts.ItemStyleOpts(color="#01B8AA"),
)
.add_yaxis(
series_name="长阳线数量",
xaxis_index=1,
yaxis_index=1,
yaxis_data=lst_bigUpBoxBar,
label_opts=opts.LabelOpts(is_show=False),
itemstyle_opts=opts.ItemStyleOpts(color="red"),
)
.add_yaxis(
series_name="长阴线数量",
xaxis_index=1,
yaxis_index=1,
yaxis_data=lst_bigDnBoxBar,
label_opts=opts.LabelOpts(is_show=False),
itemstyle_opts=opts.ItemStyleOpts(color="green"),
)
.set_global_opts(
xaxis_opts=opts.AxisOpts(name="小时/分钟",),
yaxis_opts=opts.AxisOpts(
axislabel_opts=opts.LabelOpts(is_show=True),
axisline_opts=opts.AxisLineOpts(is_show=True),
axistick_opts=opts.AxisTickOpts(is_show=True),
splitline_opts=opts.SplitLineOpts(is_show=True),
),
legend_opts=opts.LegendOpts(is_show=True,pos_bottom='0%', pos_left="center"),
)
)
gridChart = Grid()
gridChart.add(
marketSummaryBoxPlot,
grid_opts=opts.GridOpts(pos_left="10%", pos_right="8%", pos_bottom='45%'),
)
gridChart.add(
pivotCountChart,
grid_opts=opts.GridOpts(pos_left="10%", pos_right="8%", pos_top="55%"),
)
fname = '{}marketSummary{}.html'.format(cfg.PATH_ANAFILE,int_latestViewDate)
gridChart.render(fname) if renderfile else None
return gridChart
(Boxplot(init_opts=opts.InitOpts(width="1600px", height="800px")).add_xaxis(
xaxis_data=axis_data).add_yaxis(
series_name="category0",
y_axis=data[0]["boxData"],
tooltip_opts=opts.TooltipOpts(
formatter=produce_js_func("""function(param) { return [
'Experiment ' + param.name + ': ',
'upper: ' + param.data[0],
'Q1: ' + param.data[1],
'median: ' + param.data[2],
'Q3: ' + param.data[3],
'lower: ' + param.data[4]
].join('<br/>') }"""))).add_yaxis(
series_name="category1",
y_axis=data[1]["boxData"],
tooltip_opts=opts.TooltipOpts(
formatter=produce_js_func("""function(param) { return [
'Experiment ' + param.name + ': ',
'upper: ' + param.data[0],
'Q1: ' + param.data[1],
'median: ' + param.data[2],
'Q3: ' + param.data[3],
'lower: ' + param.data[4]
].join('<br/>') }"""))).
add_yaxis(series_name="category2",
y_axis=data[2]["boxData"],
tooltip_opts=opts.TooltipOpts(
formatter=produce_js_func("""function(param) { return [
'Experiment ' + param.name + ': ',
'upper: ' + param.data[0],
'Q1: ' + param.data[1],
'median: ' + param.data[2],
'Q3: ' + param.data[3],
'lower: ' + param.data[4]
].join('<br/>') }"""))).set_global_opts(
title_opts=opts.TitleOpts(title="Multiple Categories",
pos_left="center"),
legend_opts=opts.LegendOpts(pos_top="3%"),
tooltip_opts=opts.TooltipOpts(trigger="item",
axis_pointer_type="shadow"),
xaxis_opts=opts.AxisOpts(
name_gap=30,
boundary_gap=True,
splitarea_opts=opts.SplitAreaOpts(
areastyle_opts=opts.AreaStyleOpts(opacity=1)),
axislabel_opts=opts.LabelOpts(formatter="expr {value}"),
splitline_opts=opts.SplitLineOpts(is_show=False)),
yaxis_opts=opts.AxisOpts(
type_="value",
min_=-400,
max_=600,
splitarea_opts=opts.SplitAreaOpts(is_show=False)),
datazoom_opts=[
opts.DataZoomOpts(type_="inside",
range_start=0,
range_end=20),
opts.DataZoomOpts(type_="slider",
xaxis_index=0,
is_show=True)
]).render("multiple_categories.html"))
slope = "{:0.2f}".format(slope)
if slope <= str(0.1):
list_5.append(df['height'][i])
if str(0.1) < slope <= str(0.5):
list_10.append(df['height'][i])
if str(0.5) < slope <= str(0.9):
list_15.append(df['height'][i])
if str(0.9) < slope <= str(1.3):
list_20.append(df['height'][i])
if slope > str(1.3):
list_30.append(df['height'][i])
sum_list = []
sum_list.append(list_5)
sum_list.append(list_10)
sum_list.append(list_15)
sum_list.append(list_20)
sum_list.append(list_30)
c = Boxplot()
c.add_xaxis(["0-0.1", "0.1-0.5", "0.5-0.9", "0.9-1.3", "1.3以上"])
c.add_yaxis("ATLAS Elevation-Airborne LiDAR Elevation(m)",
c.prepare_data(sum_list))
c.set_global_opts(
title_opts=opts.TitleOpts(title="All data"),
xaxis_opts=opts.AxisOpts(splitline_opts=opts.SplitLineOpts(is_show=True)),
yaxis_opts=opts.AxisOpts(splitline_opts=opts.SplitLineOpts(is_show=True)),
)
c.render("boxplot.html")
def district_score_box_plot(df, district, school_name):
try:
# school_name=None
middle_score = middle_school_data4box(df)
labels = list(middle_score.columns)
c = Boxplot(init_opts=opts.InitOpts(width='1750px'))
c.add_xaxis([""])
for col in labels:
value = [list(middle_score[col].dropna())]
c.add_yaxis(col, c.prepare_data(value))
c.set_global_opts(
title_opts=opts.TitleOpts(title=f"{district}-{school_name}成绩分布"),
legend_opts=opts.LegendOpts(is_show=False),
yaxis_opts=opts.AxisOpts(min_=400, max_=750))
return c
except:
v2 = [Faker.values(450, 650)]
v1 = [Faker.values(500, 700)]
c = Boxplot(init_opts=opts.InitOpts(width='1750px'))
c.add_xaxis([''])
c.add_yaxis("中学A",
c.prepare_data(v1)).add_yaxis("中学B", c.prepare_data(v2))
c.set_global_opts(title_opts=opts.TitleOpts(
title=f"{district}初中成绩分布（示例）", subtitle="当看到此页面，说明所选区目前无数据"))
return c
)
)
scatter.render_notebook()
# %% [markdown]
# #### EffectScatter -- 涟漪散点图
effect_scatter = EffectScatter()
effect_scatter.add_xaxis(Faker.choose())
effect_scatter.add_yaxis("", Faker.values(), symbol=SymbolType.ARROW)
effect_scatter.render_notebook()
# %% [markdown]
# ### Boxplot -- 箱线图
boxplot = Boxplot()
boxplot.add_xaxis(Faker.choose())
# 计算数据的最大、最小、中位、四分位数
dt1 = boxplot.prepare_data(list(zip(*[Faker.values() for i in range(20)])))
dt2 = boxplot.prepare_data(list(zip(*[Faker.values() for i in range(20)])))
boxplot.add_yaxis("cat1", dt1)
boxplot.add_yaxis("cat2", dt2)
boxplot.render_notebook()
# %% [markdown]
# ### Polar -- 极坐标
import math
data=[]
# 生成数据，满足格式`[r, \theta]`，即可以认为角度为自变量、径向为因变量
for i in range(0, 360):