Hello. I'm Tanaka from the Information Processing Laboratory of Yamada Tanaka Shoten. I study every day what a model learns and what it can be used for in unsupervised learning.
GAN was very exciting last year and accelerated at a stretch, and this year it seems to accelerate further.
We are also considering how to use it in our lab, but decided to publish the check function of the model created in the process. I hope it helps to get an image of the Generator model. (I also use Discriminator behind the scenes)
You can try the check function by accessing the following site. It is a model that outputs the face of a woman. Since the classification of Discriminator is still sweet, strange faces are also output, but I would like to improve that in the future. When reloaded, a new image will be displayed. If you change the parameter on the left and click "Create Image", the image with the parameter change will be displayed again.
Recommended Posts