Grigoriy Livshits

Hello. I'd like to say that this repository could be a piece of gold in a DL society. Unfortunately, you have not implemented quite basic operations, like sum along axis,...

Hello. I want to convert mxnet model to pytorch. I've found out bad rename mappings in mxnet_parser.py, which is easily solved. However, it's not obvious how to make new rename...

Implementation of basic linear algebra routines

4

Hello. While trying to invert a model via your framework I've faced an absence of inverse matrix computation support (torch.inverse). I've tried to implement it by myself (via looking your...

PL reg memory consumption

4

Hello. I have the following setting: batch size 12, grad accumulation 4, image_size 128. At 5024 iterations PL reg comes in, and consumes A LOT of memory. 7 GB utilization...

Is train loop memory-efficient?

1

Hi. I've found, that you unfreeze the whole GAN, and making steps only via specific optimizer (for generator and discriminator). But when you do loss.backward, gradients are computed for the...

Hello. Are you quite sure that history of optimizer (e.g. moments) should be zeroed at the beginning of each epoch?

Training-related question

2

Hello. First of all, thx for your work and article. I have 2 questions related to the early stages of learning. 1) At 45k iterations model still outputs just the...

Mapping network mode collapse.

3

Hi. This is not an issue but a topic for discussion. I've found out that on some domains with rich and frequent patterns there is a mode collapse of a...

Artifacts

3

Hi. I wonder what are those artifacts on the generated images?  It seems like they appear on your title animation with faces, too.

Jinc function

1

Hi. I wonder why you use just the jinc(||x||), whereas in the original article they use jinc(2*fc*||x||).