-
It's hard to evaluate to the performance of a GAN. How do you even do so? Look into the literature to see what kind of stuff they're doing and why they work.
-
Generator/Discriminator loss is not particularly interpretable. If one prints out these metrics via the verbose option, it's difficult to assess whether the GAN is improving or not. I've run scenari…
-
see if it can be improved so that is general and that it is a single call, preferably in BaseGAN, and takes care if it's a cycle-consistency model or not and so on.
rn the problem is that the calcu…
-
> - “In more uncommon cases in which institutional policies do not permit the sharing of
derived data sets, synthetic data containing the same statistical properties can be generated and shared freel…
-
## 🚀 Feature
Add new audio metrics for generative audio processing
### Motivation
The evaluation of speech processing (denoising, dereverberation and in general enhancement) highly depends o…
-
https://arxiv.org/abs/1705.10843
"In unsupervised data generation tasks, besides the generation of a sample based on previous observations, one would often like to give hints to the model in order …
mrwns updated
7 years ago
-
Hello,
I've been training a Pix2Pix model for an image-to-image translation task and have reached a point where I need some advice on selecting the best epoch/weights based on the training metrics.…
-
https://arxiv.org/abs/1804.01694
> Generative Adversarial Networks (GANs) represent an attractive and novel approach to generate realistic data, such as genes, proteins, or drugs, in synthetic biol…
-
First of all, thank you very much for your great paper and code sharing. While working on my paper, I would like to measure the metric presented in the paper to compare my work with wav2lip results.
…
-
Hello,
I would like to propose adding a Generative Adversarial Network (GAN) model specifically designed for face generation to the project under SSOC'24. This model would be a valuable addition to o…