基于Basemap的带数据气泡世界地图代码优化问询

Let’s tackle your core problems—misplaced bubbles, missing country context, and unconfigured bubble labels/sizing—with a hands-on, working solution that aligns with your target output.

Key Problems in Your Original Code #

• You were using Res and Sea columns as coordinates instead of actual latitude/longitude values (that’s why bubbles were in the wrong spot!)
• No logic to scale bubbles by conversion rate or add descriptive labels
• Missing country borders to ground the map and make locations clear

Modified Full Code #

First, install any missing dependencies if needed:

pip install pycountry pandas matplotlib basemap

Here’s the revised code with all your requirements met:

import pandas as pd
from mpl_toolkits.basemap import Basemap
import matplotlib.pyplot as plt
import pycountry

# Load your dataset (replace with your actual data loading logic)
conversion_comparison = pd.DataFrame([
    {"Country": "Spain", "Sea": 6179, "Res": 85, "ConvRate(%)": 1.38, "Country_codes": "ES"},
    {"Country": "United Kingdom", "Sea": 495, "Res": 99, "ConvRate(%)": 2.00, "Country_codes": "GB"},
    {"Country": "France", "Sea": 473, "Res": 12, "ConvRate(%)": 2.55, "Country_codes": "FR"},
    {"Country": "United States", "Sea": 442, "Res": 7.8, "ConvRate(%)": 1.76, "Country_codes": "US"},
    {"Country": "Italy", "Sea": 358, "Res": 7.4, "ConvRate(%)": 2.07, "Country_codes": "IT"},
    {"Country": "Germany", "Sea": 153, "Res": 3.3, "ConvRate(%)": 2.15, "Country_codes": "DE"},
    {"Country": "Argentina", "Sea": 135, "Res": 1.9, "ConvRate(%)": 1.41, "Country_codes": "AR"},
    {"Country": "Ireland", "Sea": 132, "Res": 3.3, "ConvRate(%)": 2.49, "Country_codes": "IE"},
    {"Country": "Belgium", "Sea": 122, "Res": 4.3, "ConvRate(%)": 3.51, "Country_codes": "BE"},
    {"Country": "Israel", "Sea": 109, "Res": 2.2, "ConvRate(%)": 1.82, "Country_codes": "IL"},
])

# Function to fetch country center coordinates using pycountry + Basemap geocoding
def get_country_coords(country_name):
    try:
        country = pycountry.countries.lookup(country_name)
        return m.geocode(country.name, returnxy=True)
    except:
        return None, None

# Set up high-resolution figure
my_dpi = 100
plt.figure(figsize=(26, 18), dpi=my_dpi)

# Initialize Basemap with full world coverage and clean projection
m = Basemap(projection='mill', llcrnrlon=-180, llcrnrlat=-65, urcrnrlon=180, urcrnrlat=80)
m.drawmapboundary(fill_color='#A6CAE0', linewidth=0)
m.fillcontinents(color='grey', alpha=0.3)
m.drawcoastlines(linewidth=0.2, color="white")
m.drawcountries(linewidth=0.3, color="white")  # Add country borders for clarity

# Add latitude/longitude columns to your dataset
conversion_comparison['lon'], conversion_comparison['lat'] = zip(*conversion_comparison['Country'].apply(get_country_coords))
conversion_comparison = conversion_comparison.dropna(subset=['lon', 'lat'])  # Drop any invalid entries

# Convert geographic coordinates to map projection coordinates
x, y = m(conversion_comparison['lon'].values, conversion_comparison['lat'].values)

# Plot bubbles sized by conversion rate (scale factor adjusted for visibility)
bubble_size = conversion_comparison['ConvRate(%)'] * 600
scatter = m.scatter(x, y, s=bubble_size, alpha=0.7, c='#ff7f0e', edgecolor='white', linewidth=1)

# Add labels with country code + conversion rate
for idx, row in conversion_comparison.iterrows():
    plt.text(row['lon'] + 2, row['lat'], 
             f"{row['Country_codes']}\n{row['ConvRate(%)']}%", 
             fontsize=10, ha='left', va='center', 
             bbox=dict(facecolor='white', alpha=0.8, pad=2))

# Final layout adjustments
plt.title('Country Conversion Rates', fontsize=16, pad=20)
plt.tight_layout()
plt.show()

What Changed & Why #

1. Coordinate Fix: We added a function to fetch each country’s latitude/longitude, ensuring bubbles land exactly on their respective countries. If you don’t want to use pycountry, you can manually add lat/lon columns to your dataset for precise control.
2. Bubble Sizing: The s parameter now uses ConvRate(%) multiplied by a scale factor (600) to make higher conversion rates appear as larger bubbles.
3. Clear Labels: We loop through each data point to add a readable label with the country code and conversion rate, using a semi-transparent background to avoid clashing with the map.
4. Map Improvements:
   • Switched to the mill projection for better world coverage
   • Added drawcountries() to show distinct country borders
   • Adjusted DPI for sharper, higher-resolution output

Alternative: Hardcode Coordinates #

If you prefer to avoid external libraries, manually add these lat/lon columns to your dataframe:

conversion_comparison['lat'] = [40.4637, 55.3781, 46.2276, 37.0902, 41.8719, 51.1657, -38.4161, 53.4129, 50.5039, 31.0461]
conversion_comparison['lon'] = [-3.7492, -3.4360, 2.2137, -95.7129, 12.5674, 10.4515, -63.6167, -8.2439, 4.4699, 34.8516]

内容的提问来源于stack exchange，提问作者punit kumar Sharma