From 7cd0a8f7dbcf9f4cf55728b4db286b31248b0dc8 Mon Sep 17 00:00:00 2001
From: LufsX <lufsx@outlook.com>
Date: Tue, 15 Oct 2024 20:43:50 +0800
Subject: [PATCH] use from import single function

---
 function.py | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/function.py b/function.py
index ea64374..63f8eeb 100644
--- a/function.py
+++ b/function.py
@@ -1,6 +1,7 @@
 # function.py
-import numpy as np
 import time
+from numpy.linalg import inv
+from numpy import sum, argpartition, bincount, argmax, cov, loadtxt
 
 """
 挑战说明:
@@ -55,29 +56,29 @@ def classify0(inX, dataSet, labels, k=8, _cache={"VI": None}):
     # 管它三七二十一，直接用缓存
     if _cache["VI"] is None:
         # 缓存数据集的协方差矩阵的逆
-        _cache["VI"] = np.linalg.inv(np.cov(dataSet.T))
+        _cache["VI"] = inv(cov(dataSet.T))
 
     # 计算 inX 与数据集的差值
     diff = dataSet - inX
 
     # 使用矩阵运算计算马氏距离
-    distances = np.sum(diff @ _cache["VI"] * diff, axis=1)
+    distances = sum(diff @ _cache["VI"] * diff, axis=1)
 
     # 对距离排序，获取前 k 个最近的样本索引
-    sortedDistIndices = np.argpartition(distances, 8)[:8]
+    sortedDistIndices = argpartition(distances, 8)[:8]
 
     # 统计前 k 个样本的标签
-    classCount = np.bincount(labels[sortedDistIndices])
+    classCount = bincount(labels[sortedDistIndices])
 
     # 返回出现次数最多的标签
-    return np.argmax(classCount)
+    return argmax(classCount)
 
 
 def load_data(file_path):
     """
     加载 TSV 格式数据文件。
     """
-    data = np.loadtxt(file_path, delimiter="\t")
+    data = loadtxt(file_path, delimiter="\t")
     features = data[:, :-1]
     labels = data[:, -1].astype(int)
     return features, labels