本地计算商户之间的相似度矩阵

models/recall/itemCF/calculate_co_occurrence_matrix.py

@@ -3,6 +3,7 @@ import numpy as np
 
 from itertools import combinations
 from dao.mysql_client import Mysql
+from tqdm import tqdm
 
 def load_data_from_dataset():
     """从数据库中读取数据"""
@@ -44,9 +45,31 @@ def build_co_occurence_matrix(order_data):
             # 更新共现矩阵
             co_occurrence_matrix[idx1, idx2] += 1
             co_occurrence_matrix[idx2, idx1] += 1
-    return co_occurrence_matrix, shops
+    return co_occurrence_matrix, shops, shops_to_index
 
-def save_co_occurrence_matrix(matrix, shops, save_path):
+def calculate_similarity_matrix(co_occurrence_matrix, order_data, shops_to_index):
+    """
+    使用向量计算商铺之间的相似度矩阵
+    """
+    # 计算每个商铺售卖品规的总次数
+    shop_counts = order_data.groupby("BB_RETAIL_CUSTOMER_CODE").size()
+    num_shops = len(shops_to_index)
+    
+    # 将商户售卖次数转换为数组
+    counts = np.array([shop_counts[shop] for shop in shops_to_index.keys()])
+    
+    # 计算分母部分 (sqrt(count_i * count_j))
+    denominator = np.sqrt(np.outer(counts, counts))
+    
+    # 计算相似度矩阵
+    similarity_matrix = co_occurrence_matrix / denominator
+    
+    # 将对角线设置为1
+    np.fill_diagonal(similarity_matrix, 1.0)
+    
+    return similarity_matrix
+
+def save_matrix(matrix, shops, save_path):
     """
     保存共现矩阵
     """
@@ -54,10 +77,13 @@ def save_co_occurrence_matrix(matrix, shops, save_path):
     matrix_df.to_csv(save_path, index=True, encoding="utf-8")
     
 if __name__ == "__main__":
-    save_path = "./models/recall/itemCF/matrix/occurrence.csv"
+    co_occurrence_save_path = "./models/recall/itemCF/matrix/occurrence.csv"
+    similarity_matrix_save_path = "./models/recall/itemCF/matrix/similarity.csv"
     order_data = load_data_from_dataset()
     
-    co_occurrence_matrix, shops = build_co_occurence_matrix(order_data)
+    co_occurrence_matrix, shops, shops_to_index = build_co_occurence_matrix(order_data)
     
-    save_co_occurrence_matrix(co_occurrence_matrix, shops, save_path)
+    # save_matrix(co_occurrence_matrix, shops, co_occurrence_save_path)
+    similarity_matrix = calculate_similarity_matrix(co_occurrence_matrix, order_data, shops_to_index)
+    save_matrix(similarity_matrix, shops, similarity_matrix_save_path)