在滚动窗口中绘制快速传入数据

Question

我尝试在滚动窗口中使用 matplotlib 绘制快速数据（每秒 1-10 个数据点）。

我喜欢熊猫，因为它很简单。

我的问题是：

1. 我的框架是否高效（关于使用 pandas，matplotlib 的“plt.draw”）？

2. 如果循环运行了 1'000'000'000 次，数据框（self.df）是否会变得太大 --> 是否最好在某个时候删除数据收集器然后启动有一个空的数据框？但是滚动窗口的连续性呢？

3. 该示例在一段时间后开始变得非常缓慢。是因为绘图效率低还是因为内存使用效率低

4. 控制台显示警告：

MatplotlibDeprecationWarning：使用默认事件循环直到函数 特定于这个 GUI 实现了 warnings.warn(str, mplDeprecation)

我必须照顾它吗？

提前致谢

这是我目前所拥有的：

import numpy as np
import matplotlib.pyplot as plt; plt.ion()
import pandas as pd
import matplotlib
import matplotlib.gridspec as gridspec
plt.style.use('ggplot')
from pathlib import Path
import datetime
import matplotlib


class tradeScreen(object):

    def __init__(self):
        self.rollingWindow = 100
        self.df = pd.DataFrame(dict(time=np.NaN, bid=np.NaN, ask=np.NaN, limitBuy=np.NaN, limitSell=np.NaN, stopLoss=np.NaN), index=np.arange(self.rollingWindow))
        self.df['time'] = pd.to_datetime(self.df['time'])  # format 'time' as datetime object

        # initialise plot and line
        plt.figure()
        G = gridspec.GridSpec(2, 1)

        self.axes_1 = plt.subplot(G[0, :])
        self.axes_1.set_ylabel('First Panel')

        self.axes_2 = plt.subplot(G[1, :])
        self.axes_2.set_ylabel('Second Panel')

        self.line1, = self.axes_1.plot(self.df['time'], self.df['bid'])
        self.line2, = self.axes_1.plot(self.df['time'], self.df['ask'])
        self.line3, = self.axes_1.plot(self.df['time'], self.df['limitBuy'])
        self.line4, = self.axes_1.plot(self.df['time'], self.df['limitSell'])
        self.line5, = self.axes_2.plot(self.df['time'], self.df['stopLoss'])


    def plotter(self, tick, i):

        df = self.df
        rollingWindow = self.rollingWindow

        current_time = pd.datetime.now()

        df.loc[i, "bid"] = tick["bid"].values.item(0)
        df.loc[i, "ask"] = tick["ask"].values.item(0)
        df.loc[i, "limitBuy"] = tick["limitBuy"].values.item(0)
        df.loc[i, "limitSell"] = tick["limitSell"].values.item(0)
        df.loc[i, "stopLoss"] = tick["stopLoss"].values.item(0)
        df.loc[i, "time"] = current_time

        self.line1.set_data(pd.to_datetime(df['time'][:i].tail(rollingWindow)), df['bid'][:i].tail(rollingWindow))
        self.line2.set_data(pd.to_datetime(df['time'][:i].tail(rollingWindow)), df['ask'][:i].tail(rollingWindow))
        self.line3.set_data(pd.to_datetime(df['time'][:i].tail(rollingWindow)), df['limitBuy'][:i].tail(rollingWindow))
        self.line4.set_data(pd.to_datetime(df['time'][:i].tail(rollingWindow)), df['limitSell'][:i].tail(rollingWindow))
        self.line5.set_data(pd.to_datetime(df['time'][:i].tail(rollingWindow)), df['stopLoss'][:i].tail(rollingWindow))

        self.axes_1.autoscale_view(True, True, True)
        self.axes_1.relim()
        self.axes_2.autoscale_view(True, True, True)
        self.axes_2.relim()

        plt.draw()
        plt.pause(0.00000000000000001)


p = tradeScreen()
i = 0
for i in np.arange(300):

    # generate random data point
    t = pd.datetime.now()
    bid = np.random.rand()
    ask = np.random.rand()
    limitBuy = np.random.rand()
    limitSell = np.random.rand()
    stopLoss = np.random.rand()

    tick = pd.DataFrame(dict(time=t, bid=bid, ask=ask, limitBuy=limitBuy, limitSell=limitSell, stopLoss=stopLoss),
                        index=np.arange(1))

    p.plotter(tick, i)
    i += 1

Answer 1

您可能想查看Animation 模块以在matplotlib 中制作动画。解释见此链接：http://jakevdp.github.io/blog/2012/08/18/matplotlib-animation-tutorial/

对于后代，该页面上的第一个示例给出了Animation 系统的要点：

"""
Matplotlib Animation Example

author: Jake Vanderplas
email: vanderplas@astro.washington.edu
website: http://jakevdp.github.com
license: BSD
Please feel free to use and modify this, but keep the above information. Thanks!
"""

import numpy as np
from matplotlib import pyplot as plt
from matplotlib import animation

# First set up the figure, the axis, and the plot element we want to animate
fig = plt.figure()
ax = plt.axes(xlim=(0, 2), ylim=(-2, 2))
line, = ax.plot([], [], lw=2)

# initialization function: plot the background of each frame
def init():
    line.set_data([], [])
    return line,

# animation function.  This is called sequentially
def animate(i):
    x = np.linspace(0, 2, 1000)
    y = np.sin(2 * np.pi * (x - 0.01 * i))
    line.set_data(x, y)
    return line,

# call the animator.  blit=True means only re-draw the parts that have changed.
anim = animation.FuncAnimation(fig, animate, init_func=init,
                               frames=200, interval=20, blit=True)

使用 blitting（最后一行的blit=True）可能会提高您的性能。 但是，AFAIK matplotlib 的设计并未考虑性能，您可以查看其他更适合此任务的库，例如 PyQtGraph