降维案例 +总结

# 1、获取数据
csv_1 = pd.read_csv(r"C:\Users\v_jaketan\Desktop\1.csv")
csv_2 = pd.read_csv(r"C:\Users\v_jaketan\Desktop\2.csv")
csv_3 = pd.read_csv(r"C:\Users\v_jaketan\Desktop\3.csv")
csv_4 = pd.read_csv(r"C:\Users\v_jaketan\Desktop\4.csv")
# 2、合并表
table_1 = pd.merge(csv_1,csv_2,on=["product_id","product_id"])
table_2 = pd.merge(table_1,csv_4,on=["aisle_id","aisle_id"])
table_3 = pd.merge(table_2,csv_3,on = ["order_id","order_id"])
# 找到user_id和aisle之间关系(交叉表)
table = pd.crosstab(table_3["user_id"],table_3["aisle"])
print(table)
print(table.shape)
# 3、PCA降维
pca = PCA(n_components=0.95)
data  = pca.fit_transform(table)
print(data)
print(data.shape)

#输出结果为:

我写的例子写的很少，但还是降了一个特征。

 

总结: