[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$f5HB6nyI__KuDll04FRttpgvkutCG_gkCmNv4YryjfUg":3},{"answer":4,"createTime":5,"id":6,"options":7,"origin":10,"question":13,"related":14,"source":18,"type":19},[],"2025-05-03 08:15:43",1062184896,[8,9],"对","错",{"courseImg":11,"courseName":12},"https:\u002F\u002Ftihai-oss-cloud.itihey.com\u002Fimg\u002Fd51bc44a4ff46d5688314c55e83b443c.jpg","深度学习","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.( )",[15,20,25,30,35,40,50,55,60,65],{"answer":16,"createTime":5,"id":6,"options":17,"question":13,"source":18,"type":19},[],[8,9],"v2",3,{"answer":21,"createTime":5,"id":22,"options":23,"question":24,"source":18,"type":19},[],1062184898,[8,9],"Faster RCNN中生成推荐区域的算法为ROI Pooling.( )",{"answer":26,"createTime":5,"id":27,"options":28,"question":29,"source":18,"type":19},[],1062184900,[8,9],"TensorFlow中做深度学习计算的核心模块是tf.nn.( )",{"answer":31,"createTime":5,"id":32,"options":33,"question":34,"source":18,"type":19},[],1062184902,[8,9],"TensorFlow支持4种加载数据的方式.( )",{"answer":36,"createTime":5,"id":37,"options":38,"question":39,"source":18,"type":19},[],1062184903,[8,9],"在DCGAN中,判别器中使用的激活函数为LeakyReLu.( )",{"answer":41,"createTime":5,"id":42,"options":43,"question":48,"source":18,"type":49},[],1062184916,[44,45,46,47],"参数范数惩罚","提前终止训练","集成","增加模型的复杂度","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( )",0,{"answer":51,"createTime":5,"id":52,"options":53,"question":54,"source":18,"type":19},[],1062184918,[8,9],"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":18,"type":19},[],1062184920,[8,9],"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":18,"type":19},[],1062184928,[8,9],"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":18,"type":19},[],1062184942,[8,9],"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.( )"]