Chanil Kim

Is it right to recommend only p(roduct)4, p5 for "for('p3')" when I input add_set("u1", ["p1","p2","p3"]); add_set("u2", ["p3", "p4", "p5"])? I think it is some kind of bug in algorithm, because...

RAM Usage keep increasing

19

I've tested deepspeech on larger dataset. I found training task had finally killed by the OS before first epoch had finished. The reason I confirmed is keep increasing Memory(not GPU,...

When Extend Edges enabled(possibly because translucent nav bar), top layout guide which example views(button, and textfield added in sample project) constrained to is moved up along with setStatusBarHidden:YES called. This...

TF Binding NaN loss issue

2

When I switched from `tf.nn.ctc_loss` to `warpctc_tensorflow.ctc` I got NaN loss for the early moment of training even though `tf.nn.ctc_loss` could learn normally.  I wouldn't know how should I...

Question about pcen_cuda_kernel

1

Thanks you for sharing your code. When I'm reading your code, I've noticed that [`pcen_cuda_kernel`](https://github.com/daemon/pytorch-pcen/blob/master/pcen/pcen.py#L69) is missing. Is that part not implemented yet or omitted?