bokeh Hovertool工具提示显示日期作为数字

Question

在对csv进行一些转换后，我有一个数据帧：

ENTRYDATE |  TRANSACTIONS
2017/05/01     5
2017/05/02     20
..
2018/02/05     15

我将它转换为pandas中的datetime并使用

工具提示创建一个行图

import pandas as pd
from bokeh.models import ColumnDataSource,DatetimeTickFormatter, NumeralTickFormatter, HoverTool
from bokeh.plotting import figure
from bokeh.io import curdoc

df=pd.read_csv(r'C:\Users\file.csv')    
df2=df[['ENTRYDATE']] 
df2['ENTRYDATE']=pd.to_datetime(df2['ENTRYDATE'],infer_datetime_format=True)
df2=(df2.groupby([df2['ENTRYDATE'].dt.date]).size().reset_index(name='Transactions'))
#print(df2)
#print(df2.info())

source=ColumnDataSource(data=df2)

#hovertool

p=figure(plot_width=800,plot_height=500)

hover=HoverTool(tooltips=[("Date","@ENTRYDATE"),("Transactions","@Transactions")],
formatters={"ENTRYDATE":"datetime"},
mode='vline')
p.add_tools(hover)

p.scatter(x="ENTRYDATE",y="Transactions",color='blue', source=source)
p.xaxis.formatter=DatetimeTickFormatter()
p.yaxis.formatter=NumeralTickFormatter(format="0")

curdoc().add_root(p)

出于某种原因，显示ENTRYDATE的工具提示实际上并不显示日期（即2017-05-01），而是显示一些长数字（即14962752000000）。有人可以告诉我如何编辑这个以便工具提示以日期格式显示日期吗？

Answer 1

您需要在formatters dict中告诉它 CDS列名，并且还需要为该字段指定所需的实际特定日期时间格式，例如

HoverTool(tooltips=[
    ("Date", "@ENTRYDATE{%F}"),  # must specify desired format here
    ("Transactions", "@Transactions")
], formatters={"ENTRYDATE":"datetime"}, mode='vline')

Answer 2

另一种方法是将日期转换为字符串作为单独的列。从Yahoo获取一些库存数据。我有以下代码

import pandas as pd

# use bp as bokeh plot
from bokeh.plotting import figure, output_file, show
from bokeh.models import Range1d, HoverTool, ColumnDataSource


# get the datetime library for date & time calcs
from datetime import datetime, timedelta

data = pd.read_csv("DJI.csv")

data.head()

初始数据框如下图所示。

然后执行以下绘图：

# clean up the date fields to become dates

data['Date'] = pd.to_datetime(data['Date'], format='%Y-%m-%d')

# create a string column for plotting
data["DateString"] = data["Date"].dt.strftime("%Y-%m-%d")

# work on the bokeh plot
# set the data source for the plots and hover tool
cds = ColumnDataSource(data)

bp = figure(width=500, height=250, x_axis_type="datetime", sizing_mode="scale_width")
# bp.x_range = Range1d(bounds=(2015-12-31, 2018-12-28))

# make the hover tool and add it
hover = HoverTool(tooltips = [("Date","@DateString"),("Close", "@Close")])
bp.add_tools(hover)

bp.line("Date", "Close", color = "Red", alpha = 0.5, source = cds)

output_file("Timeseries.html")
show(bp)