251029更新

rfm/main.py

@@ -1,6 +1,5 @@
 # -*- coding: utf-8 -*-
 
-if __name__ == "__main__":
 
 """
 基于RFM模型生成数据分析报告
@@ -8,61 +7,42 @@ if __name__ == "__main__":
 
 # 导入模块
 
-    import pandas
-
+import statistics
 from datetime import datetime
-
 from decimal import Decimal, ROUND_HALF_UP
 
-    import statistics
-
+import pandas
 from jinja2 import Environment, FileSystemLoader
 
 from utils.client import MySQLClient
-
 from utils.pandas_extension import DrawAsHTML
 
+
 # 函数说明：根据RFM编码映射为客户分类
 def map_classification(r_encoded, f_encoded, m_encoded):
-
     # 就R、F、M指标构建独热编码并匹配客户分类
     match f"{r_encoded}{f_encoded}{m_encoded}":
-
         case "000":
-
             classification = "流失客户"
-
         case "010":
-
             classification = "一般维持客户"
-
         case "100":
-
             classification = "新客户"
-
         case "110":
-
             classification = "潜力客户"
-
         case "001":
-
             classification = "重要挽留客户"
-
         case "101":
-
             classification = "重要深耕客户"
-
         case "011":
-
             classification = "重要唤回客户"
-
         case "111":
-
             classification = "重要价值客户"
 
     # noinspection PyUnboundLocalVariable
     return classification
 
+
 print("1 加载数据集...", end="")
 
 client = MySQLClient(database="data_analysis")
@@ -109,9 +89,7 @@ if __name__ == "__main__":
         ),
         F=(
             "客户ID",
-                lambda x: Decimal(len(x)).quantize(
-                    Decimal("0"), rounding=ROUND_HALF_UP
-                ),
+            lambda x: Decimal(len(x)).quantize(Decimal("0"), rounding=ROUND_HALF_UP),
         ),
         M=(
             "交易金额",
@@ -136,15 +114,9 @@ if __name__ == "__main__":
 # R、F和M的平均数，使用STATISTICS.MEAN统计平均值，保证精度
 # noinspection PyUnresolvedReferences
 means = {
-        "R": statistics.mean(rfm["R"]).quantize(
-            Decimal("0.00"), rounding=ROUND_HALF_UP
-        ),
-        "F": statistics.mean(rfm["F"]).quantize(
-            Decimal("0.00"), rounding=ROUND_HALF_UP
-        ),
-        "M": statistics.mean(rfm["M"]).quantize(
-            Decimal("0.00"), rounding=ROUND_HALF_UP
-        ),
+    "R": statistics.mean(rfm["R"]).quantize(Decimal("0.00"), rounding=ROUND_HALF_UP),
+    "F": statistics.mean(rfm["F"]).quantize(Decimal("0.00"), rounding=ROUND_HALF_UP),
+    "M": statistics.mean(rfm["M"]).quantize(Decimal("0.00"), rounding=ROUND_HALF_UP),
 }
 
 rfm = rfm.assign(
@@ -158,9 +130,7 @@ if __name__ == "__main__":
     )
 )
 
-    dataframe = dataframe.merge(
-        right=rfm[["客户ID", "客户分类"]], on="客户ID", how="left"
-    )
+dataframe = dataframe.merge(right=rfm[["客户ID", "客户分类"]], on="客户ID", how="left")
 
 print("已完成")
 
@@ -236,9 +206,7 @@ if __name__ == "__main__":
     file_name="交易金额占比.html",
 )
 
-    report_backward = pandas.DataFrame(
-        data=[], columns=["客户分类", "窗口期", "客户数"]
-    )
+report_backward = pandas.DataFrame(data=[], columns=["客户分类", "窗口期", "客户数"])
 
 for customer_type in customer_types["客户分类"]:
 
@@ -253,9 +221,7 @@ if __name__ == "__main__":
         ).start_time.date()
 
         # 窗口期止期
-            period_end = pandas.Period(
-                value=f"2013-{month:02d}", freq="M"
-            ).end_time.date()
+        period_end = pandas.Period(value=f"2013-{month:02d}", freq="M").end_time.date()
 
         # 指定客户分类窗口期内客户数
         customer_counts = dataframe.loc[
@@ -392,7 +358,6 @@ if __name__ == "__main__":
 )
 
 with open("rfm_report.html", "w", encoding="utf8") as file:
-
     file.write(rfm_report)
 
 print("已完成")