Senior Robotics Software Engineer @ Rapyuta Robotics

Hi Taehee,

[2,1,4,4] is a vector with small numbers I chose so that calculation does not produce huge numbers. It’s just a random example to show the transposed convolution matrix calculation in that the weight layout is transposed from that of the convolution matrix. So, the actual weight values are irrevant here.

Hope that clarifies!

Hi Sai,

That’s a good question. It’s actually a convention to define such case as an infinity for KL divergence.

Reference:

https://stats.stackexchange.com/questions/362860/kl-divergence-between-which-distributions-could-be-infinity

Hello, I’m delighted to hear that you like the article. Please feel free to translate :)

If you haven’t already, please try reading my previous article on the entropy concept: Demystifying Entropy which explains it in details with diagrams to help you visualize the concept.

Hi Abeer,

Yes, you are correct.

Fixed — thanks!

Hi Jean,

I’ve attached MIT license to the repository. I hope that works for you.

You’re right. If the first try fails, the assign returns false which is propagated via the search and the some functions and ultimately returns false.

Hope that clarifies.

Hi João,

Thanks for pointing that out! I’ve fixed it and hope it’s easier to read now!

Hi vahid,

With GAN, we train the generator and the discriminator in turns.

When the generator is trained, the discriminator is used as a classifier and as such it is set to non-trainable.

When the discriminator is trained, it needs to be trainable and trained with real data…