榜单
服务器优惠
AI论文
-
Server
Server Basics Serve http:h2 Serve gRPC Serve thrift Serve Nshead Server push Avalanche Debug server issues Auto ConcurrencyLimiter Media Server json2pb Server Learn h... -
Deploy
Install TiKV Configure TiKV Benchmark and Performance Monitor and Alert Operate TiKV Learn to deploy, configure, monitor, and scale TiKV as you adopt the service into your pr... -
RPC in depth
RPC in depth New Protocol Atomic instructions IO Threading Overview Load Balancing Locality-aware Consistent Hashing Memory Management Timer keeping bthread_id RPC in ... -
Builtin Services
Builtin Services builtin services status vars connections flags rpcz cpu profiler heap profiler contention profiler Builtin Services Learn about bRPC builtin services. ... -
Builtin Services
Builtin Services builtin services status vars connections flags rpcz cpu profiler heap profiler contention profiler Builtin Services Learn about bRPC builtin services. ... -
Builtin Services
Builtin Services builtin services status vars connections flags rpcz cpu profiler heap profiler contention profiler Builtin Services Learn about bRPC builtin services. ... -
Deploy
Install TiKV Configure TiKV Benchmark and Performance Monitor and Alert Operate TiKV Learn to deploy, configure, monitor, and scale TiKV as you adopt the service into your pr... -
Builtin Services
Builtin Services builtin services status vars connections flags rpcz cpu profiler heap profiler contention profiler Builtin Services Learn about bRPC builtin services. ... -
Training Loop
Learner, Metrics, and Basic Callbacks Utils function replacing_yield [source] mk_metric [source] save_model [source] load_model [source] class Learner [source] PyT... -
梯度下降
梯度下降 梯度下降是一种非常通用的优化算法，它能够很好地解决一系列问题。梯度下降的整体思路是通过的迭代来逐渐调整参数使得损失函数达到最小值。 假设浓雾下，你迷失在了大山中，你只能感受到自己脚下的坡度。为了最快到达山底，一个最好的方法就是沿着坡度最陡的地方下山。这其实就是梯度下降所做的：它计算误差函数关于参数向量的局部梯度，同时它沿着梯度下降的方向进行下...