From 15d8469aa10ba3cedc4673c6b4aef97b442ff104 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=E6=98=8E=E8=BE=89123456?= <3165095743@qq.com>
Date: Tue, 10 Sep 2024 13:59:29 +0000
Subject: [PATCH] update optimal_knn.py.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: 明辉123456 <3165095743@qq.com>
---
 optimal_knn.py | 48 +++++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 47 insertions(+), 1 deletion(-)

diff --git a/optimal_knn.py b/optimal_knn.py
index 53200c3..0d89b1c 100644
--- a/optimal_knn.py
+++ b/optimal_knn.py
@@ -1,15 +1,61 @@
+import matplotlib.pyplot as plt
+from sklearn.datasets import load_digits
+from sklearn.model_selection import train_test_split
+from sklearn.neighbors import KNeighborsClassifier
+from sklearn.metrics import accuracy_score
+from tqdm import tqdm
+import pickle
 # TODO: 导入必要的库和模块
 
 # TODO: 加载数字数据集
+digits = load_digits()
+X = digits.data
+y = digits.target
 
 # TODO: 将数据集划分为训练集和测试集
+X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)
 
 # TODO: 初始化变量以存储最佳准确率，相应的k值和最佳knn模型
+best_accuracy = 0
+best_k = 1
+best_model = None
 
 # TODO: 初始化一个列表以存储每个k值的准确率
+accuracies = []
 
 # TODO: 尝试从1到40的k值，对于每个k值，训练knn模型，保存最佳准确率，k值和knn模型
+for k in tqdm(range(1, 41), desc="训练KNN模型"):
+    model = KNeighborsClassifier(n_neighbors=k)
+    model.fit(X_train, y_train)
+    y_pred = model.predict(X_test)
+    accuracy = accuracy_score(y_test, y_pred)
+
+    # 存储每个k值的准确率
+    accuracies.append(accuracy)
+
+    # 更新最佳模型
+    if accuracy > best_accuracy:
+        best_accuracy = accuracy
+        best_k = k
+        best_model = model
 
 # TODO: 将最佳KNN模型保存到二进制文件
+with open('best_knn_model.pkl', 'wb') as f:
+    pickle.dump(best_model, f)
+
+# TODO: 打印最佳准确率和相应的k值
+print(f"最佳k值: {best_k}, 最佳准确率: {best_accuracy}")
 
-# TODO: 打印最佳准确率和相应的k值
\ No newline at end of file
+# TODO: 保存准确率图到PDF文件
+plt.figure(figsize=(10, 6))
+plt.plot(range(1, 41), accuracies, marker='o', linestyle='-', color='b', label='Accuracy')
+plt.axvline(x=best_k, color='r', linestyle='--', label=f'Best k value: {best_k}')
+plt.text(best_k, best_accuracy, f'k={best_k}\nAccuracy={best_accuracy:.2f}',
+         verticalalignment='bottom', horizontalalignment='right', color='red')
+plt.xlabel('K Value')
+plt.ylabel('Accuracy')
+plt.title('Accuracy of different k values')
+plt.legend()
+plt.grid(True)
+plt.savefig('accuracy_plot.pdf')
+plt.show()
-- 
Gitee