1. 评估指标概述
AI模型评估是确保模型质量的关键步骤，不同类型的模型需要使用不同的评估指标。更多学习教程www.fgedu.net.cn
2. 分类模型评估
分类模型评估主要包括准确率、精确率、召回率、F1值等指标。学习交流加群风哥微信: itpux-com
import numpy as np
from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, confusion_matrix, classification_report
# 示例数据
y_true = [0, 1, 0, 1, 1, 0, 0, 1]
y_pred = [0, 1, 1, 1, 0, 0, 0, 1]
# 计算准确率
accuracy = accuracy_score(y_true, y_pred)
print(f”准确率: {accuracy:.4f}”)
# 计算精确率
precision = precision_score(y_true, y_pred)
print(f”精确率: {precision:.4f}”)
# 计算召回率
recall = recall_score(y_true, y_pred)
print(f”召回率: {recall:.4f}”)
# 计算F1值
f1 = f1_score(y_true, y_pred)
print(f”F1值: {f1:.4f}”)
# 生成混淆矩阵
cm = confusion_matrix(y_true, y_pred)
print(“混淆矩阵:”)
print(cm)
# 生成分类报告
report = classification_report(y_true, y_pred)
print(“分类报告:”)
print(report)
3. 回归模型评估
回归模型评估主要包括均方误差、均方根误差、平均绝对误差、R²值等指标。学习交流加群风哥QQ113257174
from sklearn.metrics import mean_squared_error, mean_absolute_error, r2_score
# 示例数据
y_true = [1.2, 2.1, 3.3, 4.5, 5.2]
y_pred = [1.0, 2.0, 3.5, 4.8, 5.0]
# 计算均方误差
mse = mean_squared_error(y_true, y_pred)
print(f”均方误差: {mse:.4f}”)
# 计算均方根误差
rmse = np.sqrt(mse)
print(f”均方根误差: {rmse:.4f}”)
# 计算平均绝对误差
ma
本文由风哥教程整理发布,仅用于学习测试使用,转载注明出处:http://www.fgedu.net.cn/10327.html
