子图属性错误：'AxesSubplot'对象没有'get_extent'属性。

Question

我正在尝试在matplotlib.pyplot上实现一个函数，它可以向我的地图中插入标尺和指南针。

我正在尝试将代码从“https://dev59.com/bVwY5IYBdhLWcg3wq5fl#35705477”适应于Geopandas geodatasets。

在我的尝试中，主要问题是获取每个绘图（Axes）的边界框坐标。出现的错误是：“subplot AttributeError: 'AxesSubplot' object has no attribute 'get_extent'”。

我尝试了许多方法来解决这个问题，但都没有成功（请参见附件中的代码）。

与下面的示例（附录中的代码）一样，我正在使用巴西的社会经济数据（来自IBGE-https://www.ibge.gov.br/estatisticas-novoportal/sociais/populacao/9109-projecao-da-populacao.html?=&t=downloads）。

这些社会经济数据已基于来自巴西的shapefile进行地理定位（在http://www.codegeo.com.br/2013/04/shapefiles-do-brasil-para-download.html中获取），并在下面的代码中命名为“SHP_joined”。因此，为了使其描述清楚，SHP_joined是一个geopandas Geodataframe，我正在尝试在其绘图中实现标尺和指南针。

我还展示了所需结果图像的示例。 “在此示例图像中，仍缺少标尺和指南针”

感谢您的时间，希望尽快得到回复。

`# -*- coding: utf-8 -*-
"""
Created on Fri Jul 20 14:53:26 2018

@author: Terry Brown - Adapted by Philipe Leal
"""

import os
import cartopy.crs as ccrs
from math import floor
import matplotlib.pyplot as plt
from matplotlib import patheffects
import numpy as np
import matplotlib
if os.name == 'nt':
    matplotlib.rc('font', family='Arial')
else:  
    # might need tweaking, must support black triangle for N arrow
    matplotlib.rc('font', family='DejaVu Sans')


def utm_from_lon(lat,lon):
    """

    :param float lon: longitude
    :return: UTM zone number
    :rtype: int
    """

    UTM_zone_number = np.int(np.floor( ( lon + 180 ) / 6) + 1)
    print("UTM Zone algorithm 1: ", UTM_zone_number)

    import utm


    UTM_zone_number2 = utm.latlon_to_zone_number(-14.2393, -54.39)

    print("UTM Zone algorithm 2: ", UTM_zone_number2)

    if UTM_zone_number2 == UTM_zone_number:
        print("OK: UTM algorithms are equal!")

        return UTM_zone_number

    else:
        print("UTM algorithms are different. Using library UTM instead!")
        return UTM_zone_number2

##### Caso Geopandas:


def scale_bar_geopandas(ax, Geopandas_dataset, length, location=(0.5, 0.05), linewidth=3,
              units='km', m_per_unit=1000):
    """

    https://dev59.com/bVwY5IYBdhLWcg3wq5fl#35705477
    ax is the axes to draw the scalebar on.
    proj is the projection the axes are in
    location is center of the scalebar in axis coordinates ie. 0.5 is the middle of the plot
    length is the length of the scalebar in km.
    linewidth is the thickness of the scalebar.
    units is the name of the unit
    m_per_unit is the number of meters in a unit
    """
    # find lat/lon center to find best UTM zone

    Minx, Miny, Maxx, Maxy = Geopandas_dataset.total_bounds

    Latitude_central = (Miny+ Maxy) /2.

    Longitude_central = (Minx + Maxx) /2.

    print("Latitude_central: ", Latitude_central)

    print("Longitude_central: ", Longitude_central)
    # Projection in metres

    try: 
        utm = ccrs.UTM(utm_from_lon( Latitude_central, Longitude_central))

    except:
        utm = ccrs.UTM(utm_from_lon(Latitude_central, Longitude_central),
                        southern_hemisphere=True)


    # Get the extent of the plotted area in coordinates in metres

    # find lat/lon center to find best UTM zone

    x0, x1, y0, y1 = Minx, Miny, Maxx, Maxy

    # Turn the specified scalebar location into coordinates in metres
    sbcx, sbcy = x0 + (x1 - x0) * location[0], y0 + (y1 - y0) * location[1]
    # Generate the x coordinate for the ends of the scalebar
    bar_xs = [sbcx - length * m_per_unit/2, sbcx + length * m_per_unit/2]
    # buffer for scalebar
    buffer = [patheffects.withStroke(linewidth=5, foreground="w")]
    # Plot the scalebar with buffer
    ax.plot(bar_xs, [sbcy, sbcy], transform=ax.transAxes, color='k',
        linewidth=linewidth, path_effects=buffer)
    # buffer for text
    buffer = [patheffects.withStroke(linewidth=3, foreground="w")]
    # Plot the scalebar label
    t0 = ax.text(sbcx, sbcy, str(length) + ' ' + units, transform=ax.transAxes,
        horizontalalignment='center', verticalalignment='bottom',
        path_effects=buffer, zorder=2)
    left = x0+(x1-x0)*0.05
    # Plot the N arrow
    t1 = ax.text(left, sbcy, u'\u25B2\nN', transform=ax.transAxes,
        horizontalalignment='center', verticalalignment='bottom',
        path_effects=buffer, zorder=2)
    # Plot the scalebar without buffer, in case covered by text buffer
    ax.plot(bar_xs, [sbcy, sbcy], transform=ax.transAxes, color='k',
        linewidth=linewidth, zorder=3)








###### Casos Normais:

def scale_bar(ax, proj, length, location=(0.5, 0.05), linewidth=3,
              units='km', m_per_unit=1000):
    """

    https://dev59.com/bVwY5IYBdhLWcg3wq5fl#35705477
    ax is the axes to draw the scalebar on.
    proj is the projection the axes are in
    location is center of the scalebar in axis coordinates ie. 0.5 is the middle of the plot
    length is the length of the scalebar in km.
    linewidth is the thickness of the scalebar.
    units is the name of the unit
    m_per_unit is the number of meters in a unit
    """
    # find lat/lon center to find best UTM zone
    try:
        x0, x1, y0, y1 = ax.get_extent(proj.as_geodetic())
    except:
        try:
            print("Trying to extract tje image extent by ax.get_window_extent()")
            x0, x1, y0, y1 = ax.get_window_extent(proj.as_geodetic())

        except:
            try:
                print("Trying to extract tje image extent by np.ravel(ax.get_window_extent())")
                x0, x1, y0, y1 = np.ravel(ax.get_window_extent(proj.as_geodetic()))
                print("\n\n x0, x1, y0 e y1 acquired succesfully: \n\n")
                print(x0, x1, y0, y1, "\n\n")
            except: 
                print("Error. x0, x1, y0 e y1 not extracted!")


    Latitude_central = (y0+y1)/2.

    Longitude_central = (x0+x1)/2.

    print("Latitude_central: ", Latitude_central)

    print("Longitude_central: ", Longitude_central)
    # Projection in metres

    try: 
        utm = ccrs.UTM(utm_from_lon( Latitude_central, Longitude_central))

    except:
        utm = ccrs.UTM(utm_from_lon(Latitude_central, Longitude_central),
                        southern_hemisphere=True)


    print(utm)
    # Get the extent of the plotted area in coordinates in metres

    # find lat/lon center to find best UTM zone
    try:
        x0, x1, y0, y1 = ax.get_extent(utm)
    except:
        print("Trying to extract the image extent by ax.get_window_extent()")
        try:
            x0, x1, y0, y1 = ax.get_window_extent(utm)
        except:
            try:
                print("Trying to extract the image extent by np.ravel(ax.get_window_extent())")

                x0, x1, y0, y1 = np.ravel(ax.get_window_extent(utm))
                print("\n\n x0, x1, y0 e y1 in UTM Projection acquired succesfully: \n\n")
                print(x0, x1, y0, y1, "\n\n")

            except: 
                print("Error. x0, x1, y0 e y1 not extracted!")



    # Turn the specified scalebar location into coordinates in metres
    sbcx, sbcy = x0 + (x1 - x0) * location[0], y0 + (y1 - y0) * location[1]
    # Generate the x coordinate for the ends of the scalebar
    bar_xs = [sbcx - length * m_per_unit/2, sbcx + length * m_per_unit/2]
    # buffer for scalebar
    buffer = [patheffects.withStroke(linewidth=5, foreground="w")]
    # Plot the scalebar with buffer
    ax.plot(bar_xs, [sbcy, sbcy], transform=ax.transAxes, color='k',
        linewidth=linewidth, path_effects=buffer)
    # buffer for text
    buffer = [patheffects.withStroke(linewidth=3, foreground="w")]
    # Plot the scalebar label
    t0 = ax.text(sbcx, sbcy, str(length) + ' ' + units, transform=ax.transAxes,
        horizontalalignment='center', verticalalignment='bottom',
        path_effects=buffer, zorder=2)
    left = x0+(x1-x0)*0.05
    # Plot the N arrow
    t1 = ax.text(left, sbcy, u'\u25B2\nN', transform=ax.transAxes,
        horizontalalignment='center', verticalalignment='bottom',
        path_effects=buffer, zorder=2)
    # Plot the scalebar without buffer, in case covered by text buffer
    ax.plot(bar_xs, [sbcy, sbcy], transform=ax.transAxes, color='k',
        linewidth=linewidth, zorder=3)



############ Testing Data example:


import pandas as pd

import geopandas as gpd


file_name = r'C:\Doutorado\Tese\SINAN\Casos_hepatite_A_por_estado_por_ano\Por_Regioes_BR_por_Ano.xlsx'

## Fluxo temporal 1 ano em 1 ano:


df = pd.read_excel(file_name, sheet_name='prevalencias', header=[1,2])


stacked = df.stack()
stacked.reset_index(inplace=True)


stacked_keys = stacked.keys()

Keys_dict = {'level_0':'ANO', 'Ano':'REGIAO', 'REGIAO':'Prevalencias'}

stacked = stacked.rename(columns=Keys_dict)

stacked.set_index('REGIAO', inplace=True)


Keys_dict_index = {'Centro-Oeste': 'Centro Oeste'}

stacked = stacked.rename(index=Keys_dict_index)

# Filtrando apenas os anos acima de 2006:
stacked = stacked[stacked['ANO'] >= 2007]


stacked['Prevalencias_relativas_%'] = stacked['Prevalencias']/np.sum(stacked['Prevalencias'])*100


SHP_path = r'c:\Doutorado\Tese\SHP\Estados_do_Brasil\Brasil_UTF_8.shp'

SHP = gpd.read_file(SHP_path)

SHP.head()


SHP.set_index('REGIAO', inplace=True)


SHP_joined = SHP.join(stacked)

SHP_joined = SHP_joined[SHP_joined['ANO'] >=2007]


SHP_joined = SHP_joined.to_crs({'init': 'epsg:4326'}) ## WGS-84

Minx, Miny, Maxx, Maxy = SHP_joined.total_bounds

Latitude_central = (Miny+ Maxy) /2.

Longitude_central = (Minx + Maxx) /2.


Anos = np.unique(SHP_joined['ANO'])

Years = []
for Ano in Anos:
    if Ano == np.nan:
        None
    elif str(Ano) == 'nan':
        None
    else:
        Years.append(Ano)

Years = np.array(Years,np.int16) 


###### ------------------------------------------#############




fig, Ax = plt.subplots(nrows=4,ncols=3, sharex='col', sharey='row',
                       )
fig.suptitle('Prevalência da Hepatite-A por Região', fontsize=16)

# definindo Vmin e Vmax para garantir range entre todos os subplots:
    # para ajuste local por subplot, deletar Vmin e Vmax.
    # ex: https://gis.stackexchange.com/questions/273273/reducing-space-in-geopandas-and-matplotlib-pyplots-subplots
Vmin = SHP_joined['Prevalencias_relativas_%'].min()
Vmax = SHP_joined['Prevalencias_relativas_%'].max()


for i in range(len(Years)):
    Ano = Years[i]
    print(Ano)

    Axes = Ax.ravel()[i]


    SHP_joined[SHP_joined['ANO']==Ano].plot(ax=Axes,
                                            column='Prevalencias_relativas_%', 
                                            legend=False,
                                            cmap='viridis',
                                            vmin=Vmin, vmax=Vmax,
                                            label=str(Ano))


    Axes.set_aspect('equal')
    Axes.set_title(str(Ano), fontsize=8)
    Axes.grid()

    scale_bar_geopandas(Axes, SHP_joined, length=100000)


Axes11 = Ax.ravel()[11] 
Axes11.set_aspect('equal')
Axes11.grid()

cax = fig.add_axes([0.9, 0.17, 0.02, 0.65])
sm = plt.cm.ScalarMappable(cmap='viridis', norm=plt.Normalize(vmin=Vmin, vmax=Vmax))
sm._A = []
cbar = fig.colorbar(sm, cax=cax)
cbar.ax.set_title('Prevalencia\n relativa (%)')


#im = plt.gca().get_children()[0]
#cax = fig.add_axes([0.90,0.1,0.03,0.8]) 
#fig.colorbar(im, cax=cax)


fig.subplots_adjust(top=0.855,
                    bottom=0.065,
                    left=1.21e-17,
                    right=0.850,
                    hspace=0.5,
                    wspace=0.005)


scale_bar_geopandas(Axes11, SHP_joined, length=100000)
plt.show()`

- Philipe Riskalla Leal

---

set_extent是cartopy axes的一个方法，即使用projection=ccrs.xyz()创建的axes。在这里，您尝试将该方法用于普通axes，但失败了，因为之前没有给它任何ccrs投影。 - ImportanceOfBeingErnest

Answer 1

您的问题在于您创建的轴方式不正确，应该使用带有投影的轴而不是普通的matplotlib轴。您可以使用subplot_kws=参数将额外的参数传递给plt.subplots()，然后这些参数将被传递给每个单独的Axes对象。 以下内容是从您提供的答案进行了修改，其中函数scale_bar()完全相同。

import cartopy.crs as ccrs
from math import floor
import matplotlib.pyplot as plt
from matplotlib import patheffects


def scale_bar(ax, proj, length, location=(0.5, 0.05), linewidth=3,
              units='km', m_per_unit=1000):
    """
    https://dev59.com/bVwY5IYBdhLWcg3wq5fl#35705477
    (...)
    """
    (...)


fig, axs = plt.subplots(nrows=4, ncols=3, sharex='col', sharey='row',
                        subplot_kw={'projection': ccrs.Mercator()})  # <--- Here is the missing piece
fig.suptitle('Cyprus')
for ax in axs.flatten():
    ax.set_extent([31, 35.5, 34, 36], ccrs.Geodetic())
    ax.coastlines(resolution='10m')
    scale_bar(ax, ccrs.Mercator(), 100)
plt.show()

- Diziet Asahi

---

亲爱的Diziet Asahi，为什么您在ax.set_extent()中使用ccrs.Geodetic()而不是使用常见的投影ccrs.Mercator()？ - Philipe Riskalla Leal

我刚刚从提供的链接中复制了代码，我不知道其中的区别，因此无法对其正确性发表评论。但是你说你正在尝试将代码适应到你的情况，所以我只是将其调整为多个子图。 - Diziet Asahi