def show(x: list, y: list, title: str = "键盘键入统计"):
bar = Bar(init_opts=opts.InitOpts(theme=ThemeType.WONDERLAND))
bar.add_xaxis(x)
bar.add_yaxis("", y, category_gap="60%")
bar.set_series_opts(
itemstyle_opts={
"normal": {
"color":
JsCode("""new echarts.graphic.LinearGradient(0, 0, 0, 1, [{
offset: 0,
color: 'rgba(0, 244, 255, 1)'
}, {
offset: 1,
color: 'rgba(250, 177, 160, 1)'
}], false)"""),
"barBorderRadius": [30, 30, 30, 30],
"shadowColor":
"rgb(0, 160, 221)",
}
})
bar.set_global_opts(
title_opts=opts.TitleOpts(title=title),
datazoom_opts=[opts.DataZoomOpts(),
opts.DataZoomOpts(type_="inside")],
)
bar.load_javascript()
bar.render_notebook()
return Markup(bar.render_embed())
def show_num(df):
df_time = df.groupby(['times']).size()
values = df_time.values.tolist()
index = df_time.index.tolist()
bar = Bar()
bar.add_xaxis(index)
bar.add_yaxis("数量 & 时间", values)
bar.set_global_opts(
xaxis_opts=opts.AxisOpts(name="评论日期", axislabel_opts={"rotate": 30}))
bar.render_notebook()
def show_pic_two():
data = df.loc[:, 'city'].value_counts().sort_values(ascending=False)
citys = list(data.index)[:15]
city_count = list(data)[:15]
bar = Bar()
bar.add_xaxis(citys)
bar.add_yaxis("Top 15", city_count)
bar.set_global_opts(title_opts=opts.TitleOpts(title=""))
bar.render_notebook()
def show_star(df):
df_time = df.groupby(['times']).size()
dic = {}
for k in df_time.index:
stars = df.loc[df['times'] == str(k), 'stars']
stars = list(map(int, stars))
dic[k] = round(sum(stars) / len(stars), 2)
bar_star = Bar()
bar_star.add_xaxis([x for x in dic.keys()])
bar_star.add_yaxis("星级 & 时间", [x for x in dic.values()])
bar_star.set_global_opts(
xaxis_opts=opts.AxisOpts(name="评论日期", axislabel_opts={"rotate": 30}))
bar_star.render_notebook()
def vtradevolume(cftable, freq="D", rendered=True):
"""
aid function on visualization of trade summary
:param cftable: cftable (pandas.DataFrame) with at least date and cash columns
:param freq: one character string, frequency label, now supporting D for date,
W for week and M for month, namely the trade volume is shown based on the time unit
:returns: the Bar object
"""
### WARN: datazoom and time conflict, sliding till 1970..., need further look into pyeacharts
### very unsatisfied about current visualize effect, and it seems the order of add and set option matters a lot
if freq == "D":
datedata = [d.to_pydatetime() for d in cftable["date"]]
selldata = [[row["date"].to_pydatetime(), row["cash"]]
for _, row in cftable.iterrows() if row["cash"] > 0]
buydata = [[row["date"].to_pydatetime(), row["cash"]]
for _, row in cftable.iterrows() if row["cash"] < 0]
elif freq == "W":
cfmerge = cftable.groupby(
[cftable["date"].dt.year, cftable["date"].dt.week])["cash"].sum()
datedata = [
dt.datetime.strptime(str(a) + "4", "(%Y, %W)%w")
for a, _ in cfmerge.iteritems()
]
selldata = [[dt.datetime.strptime(str(a) + "4", "(%Y, %W)%w"), b]
for a, b in cfmerge.iteritems() if b > 0]
buydata = [[dt.datetime.strptime(str(a) + "4", "(%Y, %W)%w"), b]
for a, b in cfmerge.iteritems() if b < 0]
elif freq == "M":
cfmerge = cftable.groupby(
[cftable["date"].dt.year, cftable["date"].dt.month])["cash"].sum()
datedata = [
dt.datetime.strptime(str(a) + "15", "(%Y, %m)%d")
for a, _ in cfmerge.iteritems()
]
selldata = [[dt.datetime.strptime(str(a) + "15", "(%Y, %m)%d"), b]
for a, b in cfmerge.iteritems() if b > 0]
buydata = [[dt.datetime.strptime(str(a) + "15", "(%Y, %m)%d"), b]
for a, b in cfmerge.iteritems() if b < 0]
else:
raise ParserFailure("no such freq tag supporting")
buydata = [[d, round(x, 1)] for d, x in buydata]
selldata = [[d, round(x, 1)] for d, x in selldata]
bar = Bar()
bar.add_xaxis(datedata)
# buydata should before selldata, since emptylist in the first line would make the output fig empty: may be bug in pyecharts
bar.add_yaxis(series_name="买入", yaxis_data=buydata, category_gap="90%")
bar.add_yaxis(series_name="卖出", yaxis_data=selldata, category_gap="90%")
bar.set_global_opts(
xaxis_opts=opts.AxisOpts(type_="time"),
datazoom_opts=[opts.DataZoomOpts(range_start=99, range_end=100)],
)
if rendered:
return bar.render_notebook()
else:
return bar
def plot_histogram(name, items, values, mode='html', img_path='res.png'):
""" 绘制直方图
Note:
从 pyecharts v1 版本开始，add_yaxis 需要传入两个参数，第一个参数为 str 类型
关于 snapshot: snapshot-selenium 是 pyecharts + selenium 渲染图片的扩展，使用 selenium 需要配置 browser driver。
可以参考 selenium-python 相关介绍，推荐使用 Chrome 浏览器，可以开启 headless 模式。目前支持 Chrome, Safari。
Download: https://selenium-python.readthedocs.io/installation.html#drivers
Make sure it’s in your PATH, e. g., place it in /usr/bin or /usr/local/bin.
"""
bar = Bar()
bar.add_xaxis(items)
bar.add_yaxis(name, values)
if mode == 'html':
return bar.render()
elif mode == 'notebook':
return bar.render_notebook()
elif mode == 'img':
make_snapshot(snapshot, bar.render(), img_path)
else:
raise ValueError("mode must in ['html', 'notebook', 'img']")
job = pd.DataFrame(list(job_collection.find()))
job.to_csv("job.csv", encoding='utf-8')
job_detail = pd.DataFrame(list(details_collection.find()))
job_detail.to_csv('job_detail.csv', encoding='utf-8')
# 薪资水平
salary_distribute = job['salary'].value_counts()
bar = Bar()
bar.add_xaxis(salary_distribute.index.values.tolist()[:10])
bar.add_yaxis("", salary_distribute.values.tolist()[:10])
bar.set_global_opts(
xaxis_opts=opts.AxisOpts(axislabel_opts=opts.LabelOpts(rotate=-30)),
title_opts=opts.TitleOpts(title="薪资分布", subtitle="薪资分布前10"),
)
bar.render_notebook()
# 高薪企业
height_salary = job[job['salary'] == '30-60K']
height_salary_year_edu = height_salary[['name', 'salary', 'year', 'edu']]
# 职位要求
height_salary_index = height_salary.index.values.tolist()
details = []
for index in height_salary_index:
detail = job_detail[job_detail.index == index]['details']
details.append(detail.values.tolist()[0])
print(details)
# 词云
def vtradevolume(cftable, freq="D", rendered=True):
"""
aid function on visualization of trade summary
:param cftable: cftable (pandas.DataFrame) with at least date and cash columns
:param freq: one character string, frequency label, now supporting D for date,
W for week and M for month, namely the trade volume is shown based on the time unit
:returns: the Bar object
"""
### WARN: datazoom and time conflict, sliding till 1970..., need further look into pyeacharts
startdate = cftable.iloc[0]["date"]
if freq == "D":
# datedata = [d.to_pydatetime() for d in cftable["date"]]
datedata = pd.date_range(startdate, yesterdayobj(), freq="D")
selldata = [[row["date"].to_pydatetime(), row["cash"]]
for _, row in cftable.iterrows() if row["cash"] > 0]
buydata = [[row["date"].to_pydatetime(), row["cash"]]
for _, row in cftable.iterrows() if row["cash"] < 0]
elif freq == "W":
cfmerge = cftable.groupby(
[cftable["date"].dt.year, cftable["date"].dt.week])["cash"].sum()
# datedata = [
# dt.datetime.strptime(str(a) + "4", "(%Y, %W)%w")
# for a, _ in cfmerge.iteritems()
# ]
datedata = pd.date_range(startdate,
yesterdayobj() + pd.Timedelta(days=7),
freq="W-THU")
selldata = [[dt.datetime.strptime(str(a) + "4", "(%G, %V)%w"), b]
for a, b in cfmerge.iteritems() if b > 0]
buydata = [[dt.datetime.strptime(str(a) + "4", "(%G, %V)%w"), b]
for a, b in cfmerge.iteritems() if b < 0]
# %V pandas gives iso weeknumber which is different from python original %W or %U,
# see https://stackoverflow.com/questions/5882405/get-date-from-iso-week-number-in-python for more details
# python3.6+ required for %G and %V
# but now seems no equal distance between sell and buy data, no idea why
elif freq == "M":
cfmerge = cftable.groupby(
[cftable["date"].dt.year, cftable["date"].dt.month])["cash"].sum()
# datedata = [
# dt.datetime.strptime(str(a) + "15", "(%Y, %m)%d")
# for a, _ in cfmerge.iteritems()
# ]
datedata = pd.date_range(startdate,
yesterdayobj() + pd.Timedelta(days=31),
freq="MS")
selldata = [[dt.datetime.strptime(str(a) + "1", "(%Y, %m)%d"), b]
for a, b in cfmerge.iteritems() if b > 0]
buydata = [[dt.datetime.strptime(str(a) + "1", "(%Y, %m)%d"), b]
for a, b in cfmerge.iteritems() if b < 0]
else:
raise ParserFailure("no such freq tag supporting")
buydata = [[d, round(x, 1)] for d, x in buydata]
selldata = [[d, round(x, 1)] for d, x in selldata]
bar = Bar()
datedata = list(datedata)
bar.add_xaxis(xaxis_data=datedata)
# buydata should before selldata, since emptylist in the first line would make the output fig empty: may be bug in pyecharts
bar.add_yaxis(series_name="买入", yaxis_data=buydata)
bar.add_yaxis(series_name="卖出", yaxis_data=selldata)
bar.set_global_opts(
tooltip_opts=opts.TooltipOpts(
is_show=True,
trigger="axis",
trigger_on="mousemove",
axis_pointer_type="cross",
),
datazoom_opts=[opts.DataZoomOpts(range_start=90, range_end=100)],
)
if rendered:
return bar.render_notebook()
else:
return bar
# https://pyecharts.org/#/zh-cn/quickstart
# https://github.com/pyecharts/pyecharts
from pyecharts.charts import Bar
bar = Bar()
bar.add_xaxis(["衬衫", "羊毛衫", "雪纺衫", "裤子", "高跟鞋", "袜子"])
bar.add_yaxis("商家A", [5, 20, 36, 10, 75, 90])
# render 会生成本地 HTML 文件，默认会在当前目录生成 render.html 文件
# 也可以传入路径参数，如 bar.render("mycharts.html")
bar.render()
"""
bar.render_notebook()#在Notebook中进行渲染图像
#bar.render("1.html")#在本地生成静态网页
#make_snapshot(snapshot, bar.render(), "bar.png")#在本地生成图表截图
"""
# bar = (
# Bar()
# .add_xaxis(["衬衫", "羊毛衫", "雪纺衫", "裤子", "高跟鞋", "袜子"])
# .add_yaxis("商家A", [5, 20, 36, 10, 75, 90])
# )
# bar.render()
from pyecharts.charts import Bar
# In[47]:
# 条形图
bar1 = Bar(init_opts=opts.InitOpts(width='1280px', height='1000px'))
bar1.add_xaxis(province_num.index.tolist())
bar1.add_yaxis("省份", province_num.values.tolist())
bar1.set_global_opts(title_opts=opts.TitleOpts(title="调剂信息发布数省份分布"),
toolbox_opts=opts.ToolboxOpts(),
visualmap_opts=opts.VisualMapOpts(max_=110))
bar1.set_series_opts(label_opts=opts.LabelOpts(position='right')) # 标签
bar1.reversal_axis()
bar1.render_notebook()
# In[48]:
from pyecharts.charts import Map
# In[49]:
c = Map(init_opts=opts.InitOpts(width='800px', height='750px'))
c.add('',[list(z) for z in zip(province_num.index.tolist(), province_num.values.tolist())], 'china')
c.set_global_opts(title_opts=opts.TitleOpts('调剂信息省份分布地图'),
toolbox_opts=opts.ToolboxOpts(is_show=True),