PyPlot 中 x 轴和图形之间的颜色

Question

我有一个使用日期时间对象绘制 x 轴的图表，我希望能够在图表本身（y 值）和 x 轴下方着色。我发现这篇帖子 Matplotlib's fill_between doesnt work with plot_date, any alternatives? 描述了一个类似的问题，但建议的解决方案对我一点帮助都没有。

我的代码：

import matplotlib.patches as mpatches
import matplotlib.dates
import matplotlib.pyplot as plt
import numpy as np
import csv
from tkinter import *
from datetime import datetime
from tkinter import ttk

columns="YEAR,MONTH,DAY,HOUR,PREC,PET,Q,UZTWC,UZFWC,LZTWC,LZFPC,LZFSC,ADIMC,AET"
data_file="FFANA_000.csv"

UZTWC  = np.genfromtxt(data_file, 
                            delimiter=',', 
                            names=columns, 
                            skip_header=1, 
                            usecols=("UZTWC"))

list_of_datetimes = []
skipped_header = False;
with open(data_file, 'rt') as f:
    reader = csv.reader(f, delimiter=',', quoting=csv.QUOTE_NONE)
    for row in reader:
        if skipped_header:
            date_string = "%s/%s/%s %s" % (row[0].strip(), row[1].strip(), row[2].strip(), row[3].strip())
            dt = datetime.strptime(date_string, "%Y/%m/%d %H")
            list_of_datetimes.append(dt)
        skipped_header = True

dates = matplotlib.dates.date2num(list_of_datetimes)

fig = plt.figure(1)
#UZTWC
ax1 = fig.add_subplot(111)
plt.plot(dates, UZTWC, '-', color='b', lw=2)
ax1.fill_between(dates, 0, UZTWC)
fig.autofmt_xdate()
plt.title('UZTWC', fontsize=15)
plt.ylabel('MM', fontsize=10)
plt.tick_params(axis='both', which='major', labelsize=10)
plt.tick_params(axis='both', which='minor', labelsize=10)
plt.grid()

plt.show()

这会产生：

Traceback (most recent call last):
  File "test_color.py", line 36, in <module>
    ax1.fill_between(dates, 0, UZTWC)
  File "C:\Users\rbanks\AppData\Local\Programs\Python\Python35-32\lib\site-packages\matplotlib\__init__.py", line 1812, in inner
    return func(ax, *args, **kwargs)
  File "C:\Users\rbanks\AppData\Local\Programs\Python\Python35-32\lib\site-packages\matplotlib\axes\_axes.py", line 4608, in fill_between
   y2 = ma.masked_invalid(self.convert_yunits(y2))
  File "C:\Users\rbanks\AppData\Local\Programs\Python\Python35-32\lib\site-packages\numpy\ma\core.py", line 2300, in masked_invalid
    condition = ~(np.isfinite(a))
TypeError: ufunc 'isfinite' not supported for the input types, and the inputs could not be safely coerced to any supported types according to the casting rule ''safe''

似乎问题在于 fill_between 无法处理我的日期类型为“numpy.ndarray”。我是否需要将其转换为另一种数据类型才能使其正常工作？

编辑：经过更多测试，我发现即使在尝试使用 list_of_datetimes 并将我所有的日期时间转换为时间戳之后，我仍然会遇到这个确切的错误，所以我开始怀疑这是否是类型问题毕竟。

样本数据：

%YEAR,MO,DAY,HR,PREC(MM/DT),ET(MM/DT),Q(CMS), UZTWC(MM),UZFWC(MM),LZTWC(MM),LZFPC(MM),LZFSC(MM),ADIMC(MM), ET(MM/DT)
2012,   5,   1,   0,     0.000,     1.250,     0.003,     2.928,     0.000,     3.335,     4.806,     0.000,     6.669,     1.042
2012,   5,   1,   6,     0.000,     1.250,     0.003,     2.449,     0.000,     3.156,     4.798,     0.000,     6.312,     0.987
2012,   5,   1,  12,     0.000,     1.250,     0.003,     2.048,     0.000,     2.970,     4.789,     0.000,     5.940,     0.929
2012,   5,   1,  18,     0.000,     1.250,     0.003,     1.713,     0.000,     2.782,     4.781,     0.000,     5.564,     0.869
2012,   5,   2,   0,     0.000,     1.250,     0.003,     1.433,     0.000,     2.596,     4.772,     0.000,     5.192,     0.809
2012,   5,   2,   6,     0.000,     1.250,     0.003,     1.199,     0.000,     2.414,     4.764,     0.000,     4.829,     0.750

我在 Windows 10 上使用 Python 3.5.0 和 matplotlib 1.5.1，我通过 WinPython https://sourceforge.net/projects/winpython/获得了所有依赖项

Answer 1

我还没有确定你的原始代码出了什么问题，但我可以使用pandas：

import pandas as pd, matplotlib.pyplot as plt, matplotlib.dates as mdates

df = pd.read_csv('/path/to/yourfile.csv')
df['date'] = df['%YEAR'].astype(str)+'/'+df['MO'].astype(str)+'/'+df['DAY'].astype(str)
df['date'] = pd.to_datetime(df['date'])
dates = [date.to_pydatetime() for date in df['date']]

yyyy_mm_dd_format = mdates.DateFormatter('%Y-%m-%d')

plt.clf()
fig = plt.figure(1)
ax = fig.add_subplot(111)
ax.plot_date(dates,df[' UZTWC(MM)'],'-',color='b',lw=2)
ax.fill_between(dates,0,df[' UZTWC(MM)'])
ax.xaxis.set_major_formatter(yyyy_mm_dd_format)
ax.set_xlim(min(dates), max(dates))
fig.autofmt_xdate()
plt.show()