MLDS2018SPRING

MLDS2018SPRING copied to clipboard

hi,where is actor-critic code?

monkeyshichi

I would like to know why the data from the test set is used during the training phase. Shouldn't training sets be used?

Johnzu-2019

The loss function in WGAN GP seems a little bit confusing

12

I notice that your loss in WGAN GP is different from that in the original paper. From the theory of GAN, the loss should just use D and don't need...

Unispac

Train another dataset

1

Can this training work for another dataset? For instance sign langage dataset.

siryne

extract feature

1

Do you have the code to extract video features? How to achieve MSVD dataset feat/video.npy by using VGG16?

TianchengQ

data.json

3

how to obtain label.json? video caption

TianchengQ

how many hours is the training time for the PPO algorithm?

1

I have no GPU,the training process is too slow and i find the agent interacting with the environment spends much time.so I wanna know the total training time. thanks!

davinca

Wrong result.

2

When I run python main.py --train_pg the reward or Avg.reward is negative. What' s wrong with it?

zengqg