从函数调用时显示混淆矩阵

Question

我有一个从 scikit learn 导入随机森林分类器的函数，我用数据拟合它，最后我想显示准确度、kappa 和混淆矩阵。除了打印混淆矩阵之外的所有工作。我没有收到任何错误，但没有打印混淆矩阵。

我尝试调用 print(cm) 并且它可以工作，但它不能以通常的 pandas 数据框样式打印，这正是我正在寻找的。​​p>

这是代码

def rf_clf(X, y, test_size = 0.3, random_state = 42):
    """This function splits the data into train and test and fits it in a random forest classifier 
    to the data provided, analysing its errors (Accuracy and Kappa). Also as this is classification,
    the function will output a confusion matrix"""

    #Split data in train and test, as well as predictors (X) and targets, (y)
    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=test_size, random_state=random_state, stratify=y)

    #import random forest classifier
    base_model = RandomForestClassifier(random_state=random_state)

    #Train the model
    base_model.fit(X_train,y_train)

    #make predictions on test set
    y_pred=base_model.predict(X_test)

    #Print Accuracy and Kappa
    print("Accuracy:",metrics.accuracy_score(y_test, y_pred))
    print("Kappa:",metrics.cohen_kappa_score(y_test, y_pred))

    #create confusion matrix
    labs = [y_test[i][0] for i in range(len(y_test))]
    cm = pd.DataFrame(confusion_matrix(labs, y_pred))
    cm #here is the issue. Kinda works with print(cm)

Answer 1

1. 一开始就从 sklearn 导入指标。

   from sklearn import metrics
   

2. 当你想显示混淆矩阵时使用它。

   # Get and show confussion matrix
   cm = metrics.confusion_matrix(y_test, y_pred)
   print(cm)
   

有了这个，你应该在原始文本中查看混淆矩阵。

如果您想用颜色显示混淆矩阵，请按以下方式进行：

1. 导入

   from sklearn.metrics import confusion_matrix
   import pandas as pd
   import seaborn as sns; sns.set()
   

2. 这样使用：

   cm = confusion_matrix(y_test, y_pred)
   cmat_df = pd.DataFrame(cm, index=class_names, columns=class_names)
   ax = sns.heatmap(cmat_df, square=True, annot=True, cbar=False)
   ax.set_xlabel('Predicción')
   ax.set_ylabel('Real')`
   

3. 希望一切顺利！