search

lucidrains / lightweight-gan

Implementation of 'lightweight' GAN, proposed in ICLR 2021, in Pytorch. High resolution image generations that can be trained within a day or two
MIT License
1.62k stars 220 forks source link

Added Experimentation Tracking. #122

Closed hnhnarek closed 2 years ago

hnhnarek commented 2 years ago

Added Experimentation Tracking using Aim.

Now you can:

Track all the model hyperparameters and architectural choices. Track all types of losses. Filter all the experiments with respect to hyperparameters or the architecture Group and aggregate w.r.t. all the trackables to dive into granular experimentation assessment. Track the generated images to track how the model improves.

Screen Shot 2022-04-12 at 16 56 35 Screen Shot 2022-04-12 at 16 57 24
lucidrains commented 2 years ago

@hnhnarek Hey Narek! This is so awesome!

Want to write up something in the README on how to use it?

hnhnarek commented 2 years ago

@hnhnarek Hey Narek! This is so awesome!

Want to write up something in the README on how to use it?

@lucidrains Hey Phil, thanks for the suggestion! I've already made the changes you suggested.

lucidrains commented 2 years ago

@hnhnarek 🔥🔥🔥

lucidrains commented 2 years ago

@hnhnarek Thank you for this!

hnhnarek commented 2 years ago

@lucidrains thank you, too. I appreciate the quick review and response! Thanks for your time. ❤️