如何让matplotlib OHLC烛台图的y轴显示OHLC值而不是y轴光标位置

Question

我正在使用matplotlib创建一个OHLC烛台股票价格图表。我使用mpl_finance的candlestick_ohlc模块来创建图表。创建图表很简单，但是图表底部的x和y轴显示的是给定光标位置的日期和y轴值，但我希望x和y轴显示的是日期和日期的开盘，高，低，关闭(ohlc)值，而不仅仅是日期和y轴光标位置的值。quotes数据集采用元组列表的格式，其中每个元组包含一个数字形式的日期，后跟开盘、高位、低位、收盘和成交量。我尝试使用matplotlib的format_coord函数来指定OHLC值，但是我不知道如何让format_coord函数接受包含日期和相关OHLC值的列表作为输入，然后给出所需的日期和OHLC值输出。下面是我写的一些简化代码，它们展示了我的问题:下面的代码现在已经被修改，可以完全工作了：

import matplotlib.pyplot as plt
from matplotlib.dates import DateFormatter, WeekdayLocator, DayLocator, MONDAY
from mpl_finance import candlestick_ohlc
from matplotlib.dates import date2num, num2date


def ohlc_daily_date_axis():
    mondays = WeekdayLocator(MONDAY)  
    alldays = DayLocator()            
    weekFormatter = DateFormatter('%b %d %Y')  # e.g., Jan 12 2018

    quotes = [(737042.0, 2.72, 2.78, 2.6815, 2.74, 414378.0),
              (737045.0, 2.71, 2.77, 2.57, 2.63, 578841.0),
              (737046.0, 2.64, 2.64, 2.4228, 2.47, 1451450.0),
              (737047.0, 2.9, 3.15, 2.7, 2.96, 7230260.0),
              (737048.0, 2.92, 3.29, 2.67, 2.83, 2784110.0),
              (737049.0, 2.78, 2.82, 2.4701, 2.51, 822776.0),
              (737052.0, 2.56, 2.6344, 2.49, 2.5, 278883.0),
              (737054.0, 2.5, 2.619, 2.34, 2.6, 606002.0),
              (737055.0, 2.57, 2.63, 2.45, 2.57, 1295820.0),
              (737056.0, 2.57, 2.75, 2.51, 2.65, 435838.0)]

    fig, ax = plt.subplots(figsize=(18,5))
    plt.subplots_adjust(bottom=0.2)
    ax.xaxis.set_major_locator(mondays)
    ax.xaxis.set_minor_locator(alldays)
    ax.xaxis.set_major_formatter(weekFormatter)

    candlestick_ohlc(ax, quotes, width=0.6)

    ax.xaxis_date()
    ax.autoscale_view()
    plt.setp(plt.gca().get_xticklabels(), rotation=45, 
horizontalalignment='right')
    #the following line puts the ohlc data in the y axis display
    ax.format_coord = get_ohlc_from_date_xy
    # the following line puts the ohlc data in the x axis display
    #ax.fmt_xdata = get_ohlc_from_date_x

    plt.show()

def get_ohlc_from_date_x(dateasnum):
    print('dateasnum: ', int(dateasnum))
    quotes = [(737042.0, 2.72, 2.78, 2.6815, 2.74, 414378.0),
              (737045.0, 2.71, 2.77, 2.57, 2.63, 578841.0),
              (737046.0, 2.64, 2.64, 2.4228, 2.47, 1451450.0),
              (737047.0, 2.9, 3.15, 2.7, 2.96, 7230260.0),
              (737048.0, 2.92, 3.29, 2.67, 2.83, 2784110.0),
              (737049.0, 2.78, 2.82, 2.4701, 2.51, 822776.0),
              (737052.0, 2.56, 2.6344, 2.49, 2.5, 278883.0),
              (737054.0, 2.5, 2.619, 2.34, 2.6, 606002.0),
              (737055.0, 2.57, 2.63, 2.45, 2.57, 1295820.0),
              (737056.0, 2.57, 2.75, 2.51, 2.65, 435838.0)]
    for i in range(len(quotes)):
        if int(dateasnum) == quotes[i][0]:
            open = quotes[i][1]
            high = quotes[i][2]
            low = quotes[i][3]
            close = quotes[i][4]
            vol = quotes[i][5]
    dte = str(num2date(dateasnum).date())
    print('type(dte): ', type(dte))
    print('open: ', open)
    ohlc_str = dte + ' open: ' + str(open) + ' high: ' + str(high) + ' 
low: ' + str(low) + ' close: ' + str(close) + ' vol: ' + str(int(vol)) 
+ '    '

    return ohlc_str



def get_ohlc_from_date_xy(dateasnum,y):
    print('dateasnum: ', int(dateasnum))
    quotes = [(737042.0, 2.72, 2.78, 2.6815, 2.74, 414378.0),
              (737045.0, 2.71, 2.77, 2.57, 2.63, 578841.0),
              (737046.0, 2.64, 2.64, 2.4228, 2.47, 1451450.0),
              (737047.0, 2.9, 3.15, 2.7, 2.96, 7230260.0),
              (737048.0, 2.92, 3.29, 2.67, 2.83, 2784110.0),
              (737049.0, 2.78, 2.82, 2.4701, 2.51, 822776.0),
              (737052.0, 2.56, 2.6344, 2.49, 2.5, 278883.0),
              (737054.0, 2.5, 2.619, 2.34, 2.6, 606002.0),
              (737055.0, 2.57, 2.63, 2.45, 2.57, 1295820.0),
              (737056.0, 2.57, 2.75, 2.51, 2.65, 435838.0)]
    for i in range(len(quotes)):
        if int(dateasnum) == quotes[i][0]:
            open = quotes[i][1]
            high = quotes[i][2]
            low = quotes[i][3]
            close = quotes[i][4]
            vol = quotes[i][5]
    dte = str(num2date(dateasnum).date())
    #print('type(dte): ', type(dte))
    #print('open: ', open)
    ohlc_str = 'open: ' + str(open) + ' high: ' + str(high) + ' 
low: ' + str(low) + ' close: ' + str(close) + ' vol: ' + str(int(vol))

    return dte, ohlc_str



# This def does not work
def format_coord(x,y, quotes):
    for i in range(len(quotes)):
        if int(x) == quotes[i]:
            open = quotes[i][1]
            high = quotes[i][2]
            low = quotes[i][3]
            close = quotes[i][4]
            vol = quotes[i][5]
    y = 'open: ' + open # I'm just using open to simplify things
    x = DateFormatter('%b %d %Y')
    return (x,y)


if __name__ == '__main__':
    ohlc_daily_date_axis()

如果我按原样运行这段代码，我会得到以下错误(这是我在使用错误的def format_coord(x，y，quotes)方法时得到的错误)：

File "/Users/Me/Mee/python_db_programs/learn_matplotlib_test.py", line 33, 
in ohlc_daily_date_axis
    ax.format_coord = format_coord(quotes)
TypeError: format_coord() missing 2 required positional arguments: 'y' 
and 'quotes'

如果我注释掉ax.format_coord =format_coord(引号)行，那么代码可以正常运行，但是在x和y显示中没有我想要的日期和OHLC值。任何关于如何继续进行的帮助都将不胜感激。

我最终没有尝试更改y显示，而是将ohlc值添加到x显示。这意味着我将ax.format_coord =format_coord(引号)改为只对x坐标进行格式化的命令，即ax.fmt_xdata，然后编写了一个def，使用引号列表获取每个日期对应的ohlc数据：

ax.fmt_xdata = get_ohlc_from_date

而不是

ax.format_coord = format_coord(quotes)

然后添加这个def：

def get_ohlc_from_date(dateasnum):
    print('dateasnum: ', int(dateasnum))
    quotes = [(737042.0, 2.72, 2.78, 2.6815, 2.74, 414378.0),
              (737045.0, 2.71, 2.77, 2.57, 2.63, 578841.0),
              (737046.0, 2.64, 2.64, 2.4228, 2.47, 1451450.0),
              (737047.0, 2.9, 3.15, 2.7, 2.96, 7230260.0),
              (737048.0, 2.92, 3.29, 2.67, 2.83, 2784110.0),
              (737049.0, 2.78, 2.82, 2.4701, 2.51, 822776.0),
              (737052.0, 2.56, 2.6344, 2.49, 2.5, 278883.0),
              (737054.0, 2.5, 2.619, 2.34, 2.6, 606002.0),
              (737055.0, 2.57, 2.63, 2.45, 2.57, 1295820.0),
              (737056.0, 2.57, 2.75, 2.51, 2.65, 435838.0)]
    for i in range(len(quotes)):
        if int(dateasnum) == quotes[i][0]:
            open = quotes[i][1]
            high = quotes[i][2]
            low = quotes[i][3]
            close = quotes[i][4]
            vol = quotes[i][5]
    dte = str(num2date(dateasnum).date())
    print('type(dte): ', type(dte))
    print('open: ', open)
    ohlc_str = dte + ' open: ' + str(open) + ' high: ' + str(high) + ' 
low: ' + str(low) + ' close: ' + str(close) + ' vol: ' + str(int(vol)) 
+ '    '

    return ohlc_str

另外，因为我使用matplotlibs dateasnum函数，所以我也必须导入该函数：

from matplotlib.dates import num2date

虽然这不会将y轴坐标替换为OHLC值，但它在x轴和y轴显示中提供了OHLC值

在弄清楚如何将ohlc值添加到x轴显示后，我意识到用于将ohlc值添加到x轴显示的逻辑可以应用于y轴显示，从而允许在y轴参数中显示ohlc值。这是通过使用ax.format_coord = format_coord命令并创建一个新的def来完成的，该def将OHLC值分配给y轴返回值。我已经修改了我发布的原始代码，以便根据ax.format_coord = format_coord行或ax.fmt_xdata = get_ohlc_from_date行是否被注释掉来确定OHLC值是作为x轴显示的一部分还是作为y轴显示的一部分显示

Answer 1

下面的解决方案允许matplotlib OHLC烛台图表的x和y轴读数在Y轴读数中显示OHLC值，而不是Y轴光标位置。

import matplotlib.pyplot as plt
from matplotlib.dates import DateFormatter, WeekdayLocator, DayLocator, 
MONDAY
from mpl_finance import candlestick_ohlc
from matplotlib.dates import date2num, num2date


def ohlc_daily_date_axis():
    mondays = WeekdayLocator(MONDAY)  
    alldays = DayLocator()            
    weekFormatter = DateFormatter('%b %d %Y')  # e.g., Jan 12 2018

    quotes = [(737042.0, 2.72, 2.78, 2.6815, 2.74, 414378.0),
              (737045.0, 2.71, 2.77, 2.57, 2.63, 578841.0),
              (737046.0, 2.64, 2.64, 2.4228, 2.47, 1451450.0),
              (737047.0, 2.9, 3.15, 2.7, 2.96, 7230260.0),
              (737048.0, 2.92, 3.29, 2.67, 2.83, 2784110.0),
              (737049.0, 2.78, 2.82, 2.4701, 2.51, 822776.0),
              (737052.0, 2.56, 2.6344, 2.49, 2.5, 278883.0),
              (737054.0, 2.5, 2.619, 2.34, 2.6, 606002.0),
              (737055.0, 2.57, 2.63, 2.45, 2.57, 1295820.0),
              (737056.0, 2.57, 2.75, 2.51, 2.65, 435838.0)]

    fig, ax = plt.subplots(figsize=(18,5))
    plt.subplots_adjust(bottom=0.2)
    ax.xaxis.set_major_locator(mondays)
    ax.xaxis.set_minor_locator(alldays)
    ax.xaxis.set_major_formatter(weekFormatter)

    candlestick_ohlc(ax, quotes, width=0.6)

    ax.xaxis_date()
    ax.autoscale_view()
    plt.setp(plt.gca().get_xticklabels(), rotation=45, 
horizontalalignment='right')
    #the following line puts the ohlc data in the y axis display
    ax.format_coord = get_ohlc_from_date_xy
    # the following line puts the ohlc data in the x axis display
    #ax.fmt_xdata = get_ohlc_from_date_x

    plt.show()

def get_ohlc_from_date_x(dateasnum):
    print('dateasnum: ', int(dateasnum))
    quotes = [(737042.0, 2.72, 2.78, 2.6815, 2.74, 414378.0),
              (737045.0, 2.71, 2.77, 2.57, 2.63, 578841.0),
              (737046.0, 2.64, 2.64, 2.4228, 2.47, 1451450.0),
              (737047.0, 2.9, 3.15, 2.7, 2.96, 7230260.0),
              (737048.0, 2.92, 3.29, 2.67, 2.83, 2784110.0),
              (737049.0, 2.78, 2.82, 2.4701, 2.51, 822776.0),
              (737052.0, 2.56, 2.6344, 2.49, 2.5, 278883.0),
              (737054.0, 2.5, 2.619, 2.34, 2.6, 606002.0),
              (737055.0, 2.57, 2.63, 2.45, 2.57, 1295820.0),
              (737056.0, 2.57, 2.75, 2.51, 2.65, 435838.0)]
    for i in range(len(quotes)):
        if int(dateasnum) == quotes[i][0]:
            open = quotes[i][1]
            high = quotes[i][2]
            low = quotes[i][3]
            close = quotes[i][4]
            vol = quotes[i][5]
    dte = str(num2date(dateasnum).date())
    print('type(dte): ', type(dte))
    print('open: ', open)
    ohlc_str = dte + ' open: ' + str(open) + ' high: ' + str(high) + ' 
low: ' + str(low) + ' close: ' + str(close) + ' vol: ' + str(int(vol)) 
+ '    '

    return ohlc_str



def get_ohlc_from_date_xy(dateasnum,y):
    print('dateasnum: ', int(dateasnum))
    quotes = [(737042.0, 2.72, 2.78, 2.6815, 2.74, 414378.0),
              (737045.0, 2.71, 2.77, 2.57, 2.63, 578841.0),
              (737046.0, 2.64, 2.64, 2.4228, 2.47, 1451450.0),
              (737047.0, 2.9, 3.15, 2.7, 2.96, 7230260.0),
              (737048.0, 2.92, 3.29, 2.67, 2.83, 2784110.0),
              (737049.0, 2.78, 2.82, 2.4701, 2.51, 822776.0),
              (737052.0, 2.56, 2.6344, 2.49, 2.5, 278883.0),
              (737054.0, 2.5, 2.619, 2.34, 2.6, 606002.0),
              (737055.0, 2.57, 2.63, 2.45, 2.57, 1295820.0),
              (737056.0, 2.57, 2.75, 2.51, 2.65, 435838.0)]
    for i in range(len(quotes)):
        if int(dateasnum) == quotes[i][0]:
            open = quotes[i][1]
            high = quotes[i][2]
            low = quotes[i][3]
            close = quotes[i][4]
            vol = quotes[i][5]
    dte = str(num2date(dateasnum).date())
    #print('type(dte): ', type(dte))
    #print('open: ', open)
    ohlc_str = 'open: ' + str(open) + ' high: ' + str(high) + ' 
low: ' + str(low) + ' close: ' + str(close) + ' vol: ' + str(int(vol))

    return dte, ohlc_str






if __name__ == '__main__':
    ohlc_daily_date_axis()

Answer 2

你可以使用Plotly's CandleStick Chart，它内置了你想要的一切。

Here is a example

元组到数据帧的序列

为此，您需要将数据放在具有列["Date", "High", "Low", "Open", "Close"]的数据帧中，查看pandas.DataFrame.from_records将数据导入到dataframe中，因为它从一系列元组创建DataFrame对象。

其他

• 您可能需要将日期转换为datetime，请查看pandas.to_datetime将其转换为

问题的答案

从剧情的documentation

hover hover确定在悬停时显示哪些跟踪信息。如果设置了none或skip，则悬停时不会显示任何信息。但是，如果未设置任何事件，单击和悬停事件仍会触发。

同样值得一读的是：Hover Text and Formatting in Python

最终考虑事项

我知道这不是你想要的matplotlib，但我认为这是一个相关的答案。

代码

def generatePlotly(df):
    layout = go.Layout(
        plot_bgcolor="#FFF",      # Sets background color to white
        hovermode="x",
        hoverdistance=100,        # Distance to show hover label of data point
        spikedistance=1000,       # Distance to show spike
        xaxis=dict(
            title="Data",         # X Axis Title
            linecolor="#BCCCDC",  # Sets color of X-axis line
            showgrid=False,       # Removes X-axis grid lines
            showspikes=True,      # Show spike line for X-axis
            gridcolor="#BCCCDC",  # Grid color, if enabled
            # Format spike - Show a Line at the pointer
            spikethickness=2,
            spikedash="dot",
            spikecolor="#999999",
            spikemode="across",
            fixedrange=True,
            spikesnap="cursor",
        ),
        yaxis=dict(
            title="Preço (R$)",   # Y Axis Title
            linecolor="#BCCCDC",  # Sets color of Y-axis line
            showgrid=False,       # Removes Y-axis grid lines
            gridcolor="#BCCCDC",  # Grid color, if enabled
            showspikes=True,      # Show spike line for X-axis
            # Format spike - Show a Line at the pointer
            spikethickness=2,
            spikedash="dot",
            spikecolor="#999999",
            spikemode="across",
            fixedrange=True,
            side="right",
            spikesnap="cursor",
        ),
        margin=go.layout.Margin(
            l=0,  # left margin
            r=0,  # right margin
            b=0,  # bottom margin
            t=0,  # top margin
        ),
    )

    fig = go.Figure(
        data=[
            go.Candlestick(
                x=df["Date"],      # Your data
                open=df["Open"],   
                high=df["High"],
                low=df["Low"],
                close=df["Close"],
            )
        ],
        layout=layout,
    )
    # Remove rangeslider from the chart, you can just comment the next line
    fig.update_layout(xaxis_rangeslider_visible=False)

    # Legend position
    fig.update_layout(legend=dict(yanchor="top", y=0.99, xanchor="left", x=0.01)) 

    ############################################################
    # For this part look at EXTRAS at the bottom of the answer #
    ############################################################
    # build complete timepline from start date to end date
    dt_all = pd.date_range(start=df["Date"].iloc[0], end=df["Date"].iloc[-1])
    # retrieve the dates that ARE in the original datset
    dt_obs = [d.strftime("%Y-%m-%d") for d in pd.to_datetime(df["Date"])]
    # define dates with missing values
    dt_breaks = [d for d in dt_all.strftime("%Y-%m-%d").tolist() if not d in dt_obs]
    fig.update_xaxes(
        rangebreaks=[
            # dict(bounds=["sat", "mon"]),  # hide weekends
            dict(values=dt_breaks)
        ]
    )

    # Hover Distance and Hover Info
    # fig.update_layout(hoverdistance=0)
    # fig.update_traces(xaxis="x", hoverinfo="none")

    fig.show()

Extras

Plotly: How to style a plotly figure so that it doesn't display gaps for missing dates?