以下是修改和补充后的代码，实现了十折交叉验证的平均auc值和平均aoc曲线，平均分类报告以及平均混淆矩阵：
```python
from sklearn.preprocessing import MinMaxScaler
from sklearn.metrics import classification_report, confusion_matrix, roc_curve, auc
from sklearn.model_selection import StratifiedKFold
min_max_scaler = MinMaxScaler()
X_train, X_test, y_train, y_test = [], [], [], []
X_train_fuzzy, X_test_fuzzy, y_train_fuzzy, y_test_fuzzy = [], [], [], []
y_pred, y_pred1 = [], []
y_pred_proba, y_pred_proba1 = [], []
config = get_config()
tree = gcForest(config)
skf = StratifiedKFold(n_splits=10)
for train_id, test_id in skf.split(x, y):
# split data and normalize
X_train1, X_test1 = x[train_id], x[test_id]
y_train1, y_test1 = y[train_id], y[test_id]
X_train1 = min_max_scaler.fit_transform(X_train1)
X_test1 = min_max_scaler.transform(X_test1)
X_train1 = np.array(X_train1)
X_test1 = np.array(X_test1)
# train gcForest
tree.fit(X_train1, y_train1)
# predict on test set
y_pred11 = tree.predict(X_test1)
y_pred_proba11 = tree.predict_proba(X_test1)
# append predictions and test data
y_pred1.append(y_pred11)
y_pred_proba1.append(y_pred_proba11)
X_train.append(X_train1)
X_test.append(X_test1)
y_test.append(y_test1)
y_train.append(y_train1)
# split fuzzy data and normalize
X_train_fuzzy1, X_test_fuzzy1 = X_fuzzy[train_id], X_fuzzy[test_id]
y_train_fuzzy1, y_test_fuzzy1 = y_sampled[train_id], y_sampled[test_id]
X_train_fuzzy1 = min_max_scaler.fit_transform(X_train_fuzzy1)
X_test_fuzzy1 = min_max_scaler.transform(X_test_fuzzy1)
X_train_fuzzy1 = np.array(X_train_fuzzy1)
X_test_fuzzy1 = np.array(X_test_fuzzy1)
# train gcForest on fuzzy data
tree.fit(X_train_fuzzy1, y_train_fuzzy1)
# predict on fuzzy test set
y_predd = tree.predict(X_test_fuzzy1)
y_predd_proba = tree.predict_proba(X_test_fuzzy1)
# append predictions and test data
y_pred.append(y_predd)
y_pred_proba.append(y_predd_proba)
X_test_fuzzy.append(X_test_fuzzy1)
y_test_fuzzy.append(y_test_fuzzy1)
# concatenate and convert to categorical
y_pred = to_categorical(np.concatenate(y_pred), num_classes=3)
y_pred1 = to_categorical(np.concatenate(y_pred1), num_classes=3)
y_test = to_categorical(np.concatenate(y_test), num_classes=3)
y_test_fuzzy = to_categorical(np.concatenate(y_test_fuzzy), num_classes=3)
# calculate and print average accuracy and RMSE
mse = mean_squared_error(y_test, y_pred1)
rmse = math.sqrt(mse)
print('深度森林RMSE:', rmse)
print('深度森林Accuracy:', accuracy_score(y_test, y_pred1))
mse = mean_squared_error(y_test_fuzzy, y_pred)
rmse = math.sqrt(mse)
print('F深度森林RMSE:', rmse)
print('F深度森林Accuracy:', accuracy_score(y_test_fuzzy, y_pred))
mse = mean_squared_error(y_test, y_pred)
rmse = math.sqrt(mse)
print('F?深度森林RMSE:', rmse)
print('F?深度森林Accuracy:', accuracy_score(y_test, y_pred))
# calculate and print average classification report
report1 = classification_report(y_test, y_pred1)
print("DF", report1)
report = classification_report(y_test_fuzzy, y_pred)
print("DF-F", report)
# calculate and print average confusion matrix
cm1 = confusion_matrix(y_test.argmax(axis=1), y_pred1.argmax(axis=1))
cm = confusion_matrix(y_test_fuzzy.argmax(axis=1), y_pred.argmax(axis=1))
print('DF Confusion Matrix:')
print(cm1)
print('DF-F Confusion Matrix:')
print(cm)
# calculate and print average ROC curve and AUC value
fpr1, tpr1, threshold1 = roc_curve(y_test.ravel(), y_pred_proba1.ravel())
fpr, tpr, threshold = roc_curve(y_test_fuzzy.ravel(), y_pred_proba.ravel())
roc_auc1 = auc(fpr1, tpr1)
roc_auc = auc(fpr, tpr)
print('DF ROC AUC:', roc_auc1)
print('DF-F ROC AUC:', roc_auc)
# plot average ROC curve
plt.title('Receiver Operating Characteristic')
plt.plot(fpr1, tpr1, 'b', label = 'DF AUC = %0.2f' % roc_auc1)
plt.plot(fpr, tpr, 'g', label = 'DF-F AUC = %0.2f' % roc_auc)
plt.legend(loc = 'lower right')
plt.plot([0, 1], [0, 1],'r--')
plt.xlim([0, 1])
plt.ylim([0, 1])
plt.ylabel('True Positive Rate')
plt.xlabel('False Positive Rate')
plt.show()
```
扫一扫
987
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
