comrade bionic
comrade bionic
I found this problem too，`self.W_q = nn.Linear(query_size, num_hidden, bias=bias)` which means the H_in should be `query_size`, but the input tensor's shape is (batch_size, num_queries, num_hidden)，apparently the last dim is `num_hidden`，not...
I think the derivative is 2 * ((0.5*weight) - 0.8) * 0.5, that is 2 * 0.5 * ((0.5*weight) - 0.8) , so the result is 0.5*x - 0.8
the general forum for this : 2 * ((input*weight) - goal_pred) * input. in nerualnetwork people may dont care about the exactly coefficient of derivative, so just omit 2 and...
> Hmm, if `error = (input * weight - goal_pred) ** 2` > should derivative not be `2 * (input * weight - goal_pred)`? > Since in this example input...
实时预览模式下能够和阅读模式显示的效果一样。live preview下每行都显示一个滑动条实在是太难受了
> 同感 live preview 不支持整体滑动实在很难受，自动换行的又非常难看。 没办法，我现在把宽度进一步的放大了，这样代码都能在一行显示，每行一个滑动条实在是有点难受
In compute_track_metrics functions, all these metrics have been averaged on the sequence length T （pcm / T, area / T, dtw / T). but in the compute_batch_metrics, it seems that...
Hi, In the simulation dataset, for example, 944, there are some negative values in video_128.npy and video_128_noisy.npy, was an additional data enhancement operation performed? 