[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$f6kPXYudiQPGd-Q0LGTTdn6VO4ISvcrf5t7j2WIvUWG0":3},{"answer":4,"createTime":5,"id":6,"options":7,"origin":12,"question":15,"related":16,"source":24,"type":49},[],"2025-05-03 08:15:43",1062184916,[8,9,10,11],"参数范数惩罚","提前终止训练","集成","增加模型的复杂度",{"courseImg":13,"courseName":14},"https:\u002F\u002Ftihai-oss-cloud.itihey.com\u002Fimg\u002Fd51bc44a4ff46d5688314c55e83b443c.jpg","深度学习","深层人工神经网络具有很强的表示能力,但经常出现过拟合.为了提高神经网络的泛化能力,下列不能提高泛化能力的方法( )",[17,26,31,36,41,46,50,55,60,65],{"answer":18,"createTime":5,"id":19,"options":20,"question":23,"source":24,"type":25},[],1062184896,[21,22],"对","错","代码tf.nn.sparse_softmax_cross_entropy_with_logits(labels,logits)的作用是定义加了稀疏的softmax交叉熵.( )","v2",3,{"answer":27,"createTime":5,"id":28,"options":29,"question":30,"source":24,"type":25},[],1062184898,[21,22],"Faster RCNN中生成推荐区域的算法为ROI Pooling.( )",{"answer":32,"createTime":5,"id":33,"options":34,"question":35,"source":24,"type":25},[],1062184900,[21,22],"TensorFlow中做深度学习计算的核心模块是tf.nn.( )",{"answer":37,"createTime":5,"id":38,"options":39,"question":40,"source":24,"type":25},[],1062184902,[21,22],"TensorFlow支持4种加载数据的方式.( )",{"answer":42,"createTime":5,"id":43,"options":44,"question":45,"source":24,"type":25},[],1062184903,[21,22],"在DCGAN中,判别器中使用的激活函数为LeakyReLu.( )",{"answer":47,"createTime":5,"id":6,"options":48,"question":15,"source":24,"type":49},[],[8,9,10,11],0,{"answer":51,"createTime":5,"id":52,"options":53,"question":54,"source":24,"type":25},[],1062184918,[21,22],"TensorFlow提供了本地 分布式执行和分布式 本地执行两种执行方式.( )",{"answer":56,"createTime":5,"id":57,"options":58,"question":59,"source":24,"type":25},[],1062184920,[21,22],"TensorFlow中不同设备和机器间的通信都由Send和Recv节点进行,而Send和Recv使用Rendezvous机制完成数据交互Rendezvous是一个基于生产 消费者—消费 生产者模型设计的抽象类.( )",{"answer":61,"createTime":5,"id":62,"options":63,"question":64,"source":24,"type":25},[],1062184928,[21,22],"人工神经网络发展大致经历过(3)个阶段 ( )",{"answer":66,"createTime":5,"id":67,"options":68,"question":69,"source":24,"type":25},[],1062184942,[21,22],"计算图执行前的传参过程中,函数调用帧是Session和执行器进行交互的窗口.( )"]