基于RFMClass统计客户数量并绘制直方图的Python技术问询

Alright, since you’ve already imported your customer list into Python for RFM analysis and added the RFMClass field to your dataset, let’s walk through how to create a clear histogram that shows how many customers fall into each RFM class. I’ll cover two common Python visualization approaches—Matplotlib for basic control and Seaborn for more polished out-of-the-box plots.

Assuming your data is stored in a pandas DataFrame (standard for this kind of analysis in Python), here are step-by-step implementations:

Method 1: Basic Histogram with Matplotlib #

This gives you full control over every element of the plot:

1. First, import the required libraries:

import pandas as pd
import matplotlib.pyplot as plt

2. Let’s assume your DataFrame is named rfm_data. We’ll first count customers per RFM class, then plot:

# Calculate customer counts for each RFM class, sorted by class number
rfm_class_counts = rfm_data['RFMClass'].value_counts().sort_index()

# Create the histogram (bar plot, since RFMClass is categorical)
plt.figure(figsize=(10, 6))
rfm_class_counts.plot(kind='bar', color='slateblue')

# Add labels and styling for clarity
plt.title('Customer Distribution by RFM Class', fontsize=14)
plt.xlabel('RFM Class', fontsize=12)
plt.ylabel('Number of Customers', fontsize=12)
plt.xticks(rotation=0)  # Keep x-axis labels horizontal for readability
plt.grid(axis='y', linestyle='--', alpha=0.6)

# Optional: Add value labels on top of each bar
for index, count in enumerate(rfm_class_counts):
    plt.text(index, count + 2, str(count), ha='center', va='bottom')

plt.show()

Method 2: Polished Count Plot with Seaborn #

Seaborn simplifies creating aesthetically pleasing statistical plots, great for quick, clean visualizations:

1. Import the necessary libraries:

import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt

2. Use countplot to directly plot customer counts per RFM class:

plt.figure(figsize=(10, 6))

# Plot counts, ensuring RFM classes are sorted numerically
sns.countplot(
    data=rfm_data,
    x='RFMClass',
    order=sorted(rfm_data['RFMClass'].unique()),
    palette='coolwarm'
)

# Add labels and styling
plt.title('Customer Distribution by RFM Class', fontsize=14)
plt.xlabel('RFM Class', fontsize=12)
plt.ylabel('Customer Count', fontsize=12)
plt.grid(axis='y', linestyle='--', alpha=0.6)

# Optional: Add value labels
for bar in plt.gca().patches:
    height = bar.get_height()
    plt.text(
        bar.get_x() + bar.get_width()/2,
        height + 2,
        f'{height}',
        ha='center', va='bottom'
    )

plt.show()

Quick Notes #

• If your RFMClass is stored as a string (instead of integer), convert it to numeric before sorting to avoid odd ordering (e.g., "100" appearing before "20"):

  rfm_data['RFMClass'] = pd.to_numeric(rfm_data['RFMClass'])
  

• Adjust the figsize, color, and palette parameters to match your preferred style or reporting needs.

内容的提问来源于stack exchange，提问作者Ben P