[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$f_bUkBX6vfam1AN33KVpm0AZdZCoRjZFVrRXKuc8eq7Y":3},{"answer":4,"createTime":5,"id":6,"options":7,"origin":12,"question":19,"related":20,"source":24,"type":25},[],"2025-03-27 16:08:27",180502084,[8,9,10,11],"真正例(TP)","假正例(FP)","真反例(TN)","假反例(FN)",{"count":13,"courseId":14,"courseImg":15,"courseName":16,"workId":17,"workName":18},10,"f20e2428c31261e30db352a88a06084e","https:\u002F\u002Ftihai-oss-cloud.itihey.com\u002Fimg\u002F25b30343053994e8940089572d36015b.jpg","机器学习","39ecb729a4854f84a122aa7703bf1069","第二章-模型评估与选择","在评估分类模型时,若一个样本实际为正类,被预测为正类,这属于( )",[21,26,35,44,53,61,70,77,86,95],{"answer":22,"createTime":5,"id":6,"options":23,"question":19,"source":24,"type":25},[],[8,9,10,11],"v1",0,{"answer":27,"createTime":5,"id":28,"options":29,"question":34,"source":24,"type":25},[],180502085,[30,31,32,33],"准确率","召回率","梯度下降","F1 值","以下不属于性能度量的是( )",{"answer":36,"createTime":5,"id":37,"options":38,"question":43,"source":24,"type":25},[],180502086,[39,40,41,42],"1:9","3:7","7:3","9:1","当我们使用留出法将数据集划分为训练集和测试集时,通常训练集和测试集的比例为( )",{"answer":45,"createTime":5,"id":46,"options":47,"question":52,"source":24,"type":25},[],180502087,[48,49,50,51],"k 折交叉验证中,k 值越大,评估结果越稳定","留一法是 k 折交叉验证的特殊情况,其中 k 等于样本数","交叉验证法可以有效避免过拟合","分层采样可以保证每个折叠中的样本类别比例与原始数据集一致","以下关于交叉验证法的说法,错误的是( )",{"answer":54,"createTime":5,"id":55,"options":56,"question":60,"source":24,"type":25},[],180502088,[30,57,58,59],"查准率","ROC 曲线下的面积(AUC)","错误率","若分类任务中各类别的样本数量严重不均衡,以下哪种性能度量更能反映模型的性能?( )",{"answer":62,"createTime":5,"id":63,"options":64,"question":69,"source":24,"type":25},[],180502089,[65,66,67,68],"混淆矩阵只能用于二分类问题","混淆矩阵中对角线元素之和等于样本总数","混淆矩阵可以直观地展示模型的分类情况","混淆矩阵的行数和列数一定不相等","以下关于混淆矩阵的说法正确的是( )",{"answer":71,"createTime":5,"id":72,"options":73,"question":76,"source":24,"type":25},[],180502090,[74,75,31,30],"均方误差(MSE)","精确率","在回归任务中,常用的性能度量是( )",{"answer":78,"createTime":5,"id":79,"options":80,"question":85,"source":24,"type":25},[],180502091,[81,82,83,84],"0.2","0.5","0.8","0.6","假设我们有一个二分类模型,其真正例率(TPR)为 0.8,假正例率(FPR)为 0.2,那么该模型的 ROC 曲线下的面积(AUC)为( )",{"answer":87,"createTime":5,"id":88,"options":89,"question":94,"source":24,"type":25},[],180502092,[90,91,92,93],"增加训练数据","降低模型复杂度","早停法","增加模型参数","以下哪种方法不是用来缓解过拟合的( )",{"answer":96,"createTime":5,"id":97,"options":98,"question":103,"source":24,"type":25},[],180502093,[99,100,101,102],"可以根据模型在验证集上的性能来选择模型","模型选择的目标是找到在测试集上性能最好的模型","信息准则可以用于模型选择","模型选择只需要考虑模型的训练误差","以下关于模型选择的说法,错误的是( )"]