[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$fnQGs7V3jfeZIs93pKaYipmfYPRKzPMnm9D-tKVQqdqU":3},{"answer":4,"createTime":5,"id":6,"options":7,"origin":8,"question":15,"related":16,"source":27,"type":109},[],"2024-04-11 21:21:14",137904453,[],{"count":9,"courseId":10,"courseImg":11,"courseName":12,"workId":13,"workName":14},28,"8c84668bf17dad779db1804412d49527","https:\u002F\u002Ftihai-oss-cloud.itihey.com\u002Fimg\u002F48eba3395437dd881b1b9275a0ad1028.png","机器学习A","work_34147034","阶段复习","假设我们已经建立好了一个二分类模型, 输出是0或1, 初始阈值设置为0.5, 超过0.5概率估计就判别为1, 否则就判别为0 ; 如果我们现在用另一个大于0.5的阈值, 一般来说,查全率会__________",[17,29,38,47,57,65,74,82,91,100],{"answer":18,"createTime":19,"id":20,"options":21,"question":26,"source":27,"type":28},[],"2024-04-11 21:02:12",137899968,[22,23,24,25],"分类和回归是监督学习的代表","模型是通过学习算法得到的","机器学习通常解决高度不确定性和复杂性的问题","机器学习一定需要类别标记","下列说法错误的是()","v1",0,{"answer":30,"createTime":19,"id":31,"options":32,"question":37,"source":27,"type":28},[],137899969,[33,34,35,36],"销量预测","网页编写","文本分类","人脸识别","下列不属于机器学习任务的是()",{"answer":39,"createTime":19,"id":40,"options":41,"question":46,"source":27,"type":28},[],137899970,[42,43,44,45],"以上说法均正确","训练集小的时候,评估结果的偏差较大","测试集小的时候,评估结果的方差较大","留出法需要对数据集进行多次切分并将结果取平均值","对于留出法,下列说法正确的是",{"answer":48,"createTime":49,"id":50,"options":51,"question":56,"source":27,"type":28},[],"2024-04-11 21:02:13",137899971,[52,53,54,55],"测试性能随着测试集的变化而变化","很多机器学习算法本身有一定的随机性","以上均正确","测试性能不等于泛化性能","两种算法在某种度量下取得评估结果后不能直接比较以评判优劣的原因中,正确的是",{"answer":58,"createTime":49,"id":59,"options":60,"question":26,"source":27,"type":28},[],137899972,[61,62,63,64],"机器学习一般有"独立同分布"假设","机器学习在只要见过的数据上做好就行了,未见过样本上的性能不重要","一般假设拿到的所有数据都来自一个潜在的分布","学得模型适用于新样本的能力称为"泛化"能力",{"answer":66,"createTime":49,"id":67,"options":68,"question":73,"source":27,"type":28},[],137899973,[69,70,71,72],"查全率","F1度量","查准率","精度","当西瓜收购公司去瓜摊收购西瓜时既希望把好瓜都收走又保证收到的瓜中坏瓜尽可能的少,请问他应该考虑什么评价指标",{"answer":75,"createTime":49,"id":76,"options":77,"question":26,"source":27,"type":28},[],137899974,[78,79,80,81],"模型找出的规律一定是正确的","一般假设正类和反类是可交换的",""色泽"取值为"青绿",这里的"青绿"是属性值","输出是离散值的学习任务为分类任务",{"answer":83,"createTime":49,"id":84,"options":85,"question":90,"source":27,"type":28},[],137899975,[86,87,88,89],"学习过程就是为了找出数据的某种潜在规律,这个规律自身,一般称为"数据特征"","从数据中学得模型的过程称为"学习"或"训练"","学得模型对应了关于数据的某种潜在的规律,称为"假设"","训练过程中使用的数据称为"训练数据",每一个样本称为一个"训练样本",训练样本组成的集合称为"训练集"","下列有关机器学习基本术语的说法错误的是",{"answer":92,"createTime":49,"id":93,"options":94,"question":99,"source":27,"type":28},[],137899976,[95,96,97,98],"训练集 不需要","验证集 需要","训练集 需要","验证集 不需要","我们通常将数据集划分为训练集,验证集和测试集进行模型的训练,参数的验证需要在__上进行,参数确定后__重新训练模型",{"answer":101,"createTime":49,"id":102,"options":103,"question":108,"source":27,"type":28},[],137899977,[104,105,106,107],"在任何情况下,总有一个最优的学习算法","机器学习算法在学习过程中对某种类型假设的偏好,称为"归纳偏好",或简称为"偏好"","一般来说,任何一个有效的机器学习算法都有其归纳偏好",""奥卡姆剃刀"原则在某些情况下可以指导我们选择偏好","下列关于归纳偏好的说法错误的是()",2]