如果 x 轴是熊猫的日期时间索引，如何绘制多色线

问题描述：

我正在尝试使用 Pandas 系列绘制一条多色线。我知道matplotlib.collections.LineCollection这会大大提高效率。但是 LineCollection 要求线段必须是浮点型。我想使用 Pandas 的 datatime 索引作为 x 轴。

points = np.array((np.array[df_index.astype('float'), values]).T.reshape(-1,1,2))
segments = np.concatenate([points[:-1],points[1:]], axis=1)
lc = LineCollection(segments)
fig = plt.figure()
plt.gca().add_collection(lc)
plt.show()

但是图片不能让我满意。有什么解决办法吗？

解决方案 1：

要生成多色线条，您需要先将日期转换为数字，因为 matplotlib 内部仅适用于数字值。

对于 matplotlib 提供的转换matplotlib.dates.date2num。它可以理解 datetime 对象，因此您首先需要使用 将时间序列转换为 datetime series.index.to_pydatetime()，然后应用date2num。

s = pd.Series(y, index=dates)
inxval = mdates.date2num(s.index.to_pydatetime())

然后，您可以照常使用数字点，例如绘制为 Polygon 或 LineCollection[ 1 , 2 ]。

完整示例：

import pandas as pd
import matplotlib.pyplot as plt
import matplotlib.dates as mdates
import numpy as np
from matplotlib.collections import LineCollection

dates = pd.date_range("2017-01-01", "2017-06-20", freq="7D" )
y = np.cumsum(np.random.normal(size=len(dates)))

s = pd.Series(y, index=dates)

fig, ax = plt.subplots()

#convert dates to numbers first
inxval = mdates.date2num(s.index.to_pydatetime())
points = np.array([inxval, s.values]).T.reshape(-1,1,2)
segments = np.concatenate([points[:-1],points[1:]], axis=1)

lc = LineCollection(segments, cmap="plasma", linewidth=3)
# set color to date values
lc.set_array(inxval)
# note that you could also set the colors according to y values
# lc.set_array(s.values)
# add collection to axes
ax.add_collection(lc)


ax.xaxis.set_major_locator(mdates.MonthLocator())
ax.xaxis.set_minor_locator(mdates.DayLocator())
monthFmt = mdates.DateFormatter("%b")
ax.xaxis.set_major_formatter(monthFmt)
ax.autoscale_view()
plt.show()

由于人们似乎难以理解这个概念，下面是一段与上面相同的代码，但没有使用熊猫，而是使用了一个独立的颜色数组：

import matplotlib.pyplot as plt
import matplotlib.dates as mdates
import numpy as np; np.random.seed(42)
from matplotlib.collections import LineCollection

dates = np.arange("2017-01-01", "2017-06-20", dtype="datetime64[D]" )
y = np.cumsum(np.random.normal(size=len(dates)))
c = np.cumsum(np.random.normal(size=len(dates)))


fig, ax = plt.subplots()

#convert dates to numbers first
inxval = mdates.date2num(dates)
points = np.array([inxval, y]).T.reshape(-1,1,2)
segments = np.concatenate([points[:-1],points[1:]], axis=1)

lc = LineCollection(segments, cmap="plasma", linewidth=3)
# set color to date values
lc.set_array(c)
ax.add_collection(lc)

ax.xaxis_date()
ax.autoscale_view()
plt.show()

解决方案 2：

ImportanceOfBeingErnest 的答案非常好，帮我省了很多时间。我想分享一下我如何利用上面的答案，根据 Pandas DataFrame 中的信号改变颜色。

import matplotlib.dates as mdates
# import matplotlib.pyplot as plt
# import numpy as np
# import pandas as pd
from matplotlib.collections import LineCollection
from matplotlib.colors import ListedColormap, BoundaryNorm

制作测试 DataFrame

equity = pd.DataFrame(index=pd.date_range('20150701', periods=150))
equity['price'] = np.random.uniform(low=15500, high=18500, size=(150,))
equity['signal'] = 0
equity.signal[15:45] = 1
equity.signal[60:90] = -1
equity.signal[105:135] = 1

# Create a colormap for crimson, limegreen and gray and a norm to color
# signal = -1 crimson, signal = 1 limegreen, and signal = 0 lightgray
cmap = ListedColormap(['crimson', 'lightgray', 'limegreen'])
norm = BoundaryNorm([-1.5, -0.5, 0.5, 1.5], cmap.N)

# Convert dates to numbers
inxval = mdates.date2num(equity.index.to_pydatetime())

# Create a set of line segments so that we can color them individually
# This creates the points as a N x 1 x 2 array so that we can stack points
# together easily to get the segments. The segments array for line collection
# needs to be numlines x points per line x 2 (x and y)
points = np.array([inxval, equity.price.values]).T.reshape(-1,1,2)
segments = np.concatenate([points[:-1],points[1:]], axis=1)

# Create the line collection object, setting the colormapping parameters.
# Have to set the actual values used for colormapping separately.
lc = LineCollection(segments, cmap=cmap, norm=norm, linewidth=2)

# Set color using signal values
lc.set_array(equity.signal.values)

fig, ax = plt.subplots()
fig.autofmt_xdate()

# Add collection to axes
ax.add_collection(lc)

plt.xlim(equity.index.min(), equity.index.max())
plt.ylim(equity.price.min(), equity.price.max())
plt.tight_layout()

# plt.savefig('test_mline.png', dpi=150)
plt.show()