用pycharm实现它

在PyCharm中实现KNN模型通常按照以下步骤进行：

1. **环境设置**: 安装必要的库，例如scikit-learn用于KNN模型，pandas用于数据处理，matplotlib或seaborn用于数据可视化。在PyCharm的项目中，你可以通过"File" > "Project Structure" > "Libraries"添加这些库。

2. **导入模块**:

   import pandas as pd
   from sklearn.model_selection import train_test_split
   from sklearn.preprocessing import StandardScaler
   from sklearn.neighbors import KNeighborsClassifier
   from sklearn.metrics import accuracy_score, confusion_matrix
   import matplotlib.pyplot as plt

3. **加载和预处理数据**:

   # 加载数据
   wine_data = pd.read_csv('wine.csv')  # 假设数据文件名是wine.csv
   X = wine_data.drop('class', axis=1)  # 特征
   y = wine_data['class']  # 类别

   # 分割数据
   X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)

   # 归一化特征
   scaler = StandardScaler()
   X_train_scaled = scaler.fit_transform(X_train)
   X_test_scaled = scaler.transform(X_test)

4. **创建和训练模型**:

   knn = KNeighborsClassifier(n_neighbors=5)  # 假设初始尝试k=5
   knn.fit(X_train_scaled, y_train)

5. **预测和评估**:

   y_pred = knn.predict(X_test_scaled)
   accuracy = accuracy_score(y_test, y_pred)
   print(f"Accuracy: {accuracy}")

   cm = confusion_matrix(y_test, y_pred)
   print("Confusion Matrix:\n", cm)

6. **可视化的例子**:

   df_cm = pd.DataFrame(cm, index=wine_data['class'].unique(), columns=wine_data['class'].unique())
   plt.figure(figsize=(10, 7))
   sns.heatmap(df_cm, annot=True, cmap='Blues')
   plt.title("Confusion Matrix")
   plt.xlabel("Predicted")
   plt.ylabel("True");

7. **保存模型** (可选):

   joblib.dump(knn, 'knn_model.pkl')  # 如果你想保存模型以便后续使用