
人工智能
夜淡
这个作者很懒,什么都没留下…
展开
-
tensor复杂操作汇总
【代码】tensor复杂操作汇总。原创 2023-12-10 03:12:32 · 150 阅读 · 0 评论 -
python multiprocessing模块的坑
1.进程join造成死锁当代码如下,进程正常结束:from multiprocessing import Process, Queuedef f(q): q.put('X' * 60000)if __name__ == '__main__': queue = Queue() p1 = Process(target=f, args=(queue,)) p1...原创 2019-07-31 13:07:25 · 1396 阅读 · 0 评论 -
torch.nn.Sequential and torch.nn.Module
nn.Sequential:seq_net = nn.Sequential( nn.Linear(2,4), # define your network nn.Tanh(), nn.Linear(4,1))param = seq_net.parameters() # get all parametersoptim = torch.optim.SGD(param, ...原创 2019-05-05 00:48:15 · 309 阅读 · 0 评论 -
模型评估与选择
性能度量错误率:E(f;D)=1m∑i=1mI(f(xi≠yi))E(f;D)=\frac{1}{m}\sum_{i=1}^{m}\mathbb I(f(x_i\ne y_i))E(f;D)=m1i=1∑mI(f(xi̸=yi))原创 2019-07-23 19:26:37 · 522 阅读 · 0 评论 -
Monte Carlo Tree Search
The process of Upper Confidence Bounds for Trees (UCT)1. SelectionStarting at the root node R, search a leaf node L.Specifically, we use UCB(upper confidence bound) algorithm to choose the descenda...原创 2019-03-23 01:52:54 · 489 阅读 · 0 评论