Ruslan

icon indicating a website link tele.gg/russab0 icon indicating an email [email protected]

icon company Innopolis University icon location Innopolis / Kazan icon location Computer Science student at Innopolis University; Backend-developer at TXTCV; Teacher at SanakLab

Results 5 comments of Ruslan

Implementation in pytorch

There is a [GitHub repository](https://github.com/eriklindernoren/PyTorch-GAN), you may get some insights from that. But there are some differences with the original code (tt least, num_epochs and reconstruction loss differ)

How to avoid blocky reconstruction

@harshithabk any details? I have faced the same problem when trying to run the model on a different dataset. The reconstructed area is simply 16 blocks (attached the example) ![0025500](https://user-images.githubusercontent.com/16399704/102724629-f5197480-4321-11eb-9a20-db5ae9bea749.png)

Pixelwise loss in AAE

The same question about Context Encoder

dataset?

Btw, is the dataset really consists of 20 images?

Добавить "-" между автором и названием в имени файла

Есть даже соответствующий PR https://github.com/sodeprecated/yandex-music-downloader/pull/20