kunrenzhilu

Successful build but unable to get some core linuxmodule elements in output.

8

Hi, the click router doesn't compile some of the core elements, e.g. todevice.cc, tohost.cc, fromdevice.cc and etc. I confirmed it by looking at the {CLICKPATH}/linuxmodule/elements.conf, whereas other elements like touserdevice.cc...

Why do you use a separate rollout instead of directly using the Generator?

1

Since you use the generator's updated weights to update the rollout, why do you separate these two?

I use the follow code: ```python mat = np.repeat(np.repeat(np.arange(9).reshape(3,3), 3, axis=0), 3, axis=1).reshape(1,1,9,9) mat2 = torch.from_numpy(mat) P.im2col(Variable(mat2.cuda()), (3,3), (3,3), (0,0)) ``` Then it throws error: ``` --------------------------------------------------------------------------- CompileException Traceback (most...

[This line](https://github.com/github-pengge/PyTorch-progressive_growing_of_gans/blob/96813876ae869a1176e9abc3231d4fb29ab5c09c/train.py#L272) ``` python for it in range(from_it, total_it): if phase == 'stabilize': cur_level = R else: cur_level = R + total_it/float(from_it) cur_resol = 2 ** int(np.ceil(cur_level+1)) ``` is problematic....

loss got nan.

5

Train Epoch: 3 [0/60000 (0%)] KLD Loss: 2.687659 NLL Loss: 73.599564 Train Epoch: 3 [2800/60000 (21%)] KLD Loss: 2.976363 NLL Loss: 78.757454 Train Epoch: 3 [5600/60000 (43%)] KLD Loss: 2.837864...

inference for bi-variate gaussion

1

Hi I want to extend the abstractEmissions to implement a inference algorithm for bi-variate gaussian. Can you give me any references on how to do so? Or, from where you...

`rewards = _backwards_cumsum(decays * tf.sigmoid(self.d_gen_predictions), self.sequence_length)` ` self.g_loss = \ -tf.reduce_mean(tf.log(self.gen_o.stack()) * normalized_rewards)` I think the key to understand your code's reinforcement learning part is the above two lines. But,...

The code is not numerical stable, it throws runtime warnings from time to time

3

I test my model with the following setting and it throws an runtimewarning every time. I tested the model with the same setting in py2. Sometimes it's safe but sometimes...