PSNR ~27 on lego dataset
- Dominant language
- Jupyter Notebook
- Stars
- 2.8k
- Forks
- 458
- PR merge metrics
- No merged PRs in 30d
Description
Hello,
Thank you for this great work on pytorch-lightning.
I run branch dev on RTX 3090 for lego dataset, follow the default config in readme.md, but the final PSNR can only reach ~27. Can you help me improve the result?
Thanks!
Here is the tensorboard log:
Contributor guide
No contributing guide indexed for this repository
Research direction
Start with the default configuration and training instructions in readme.md, then reproduce the lego dataset run on the dev branch and inspect the attached TensorBoard log. Compare the observed ~27 PSNR result with the documented run and trace the configuration or training differences; done means the cause is identified and a reproducible correction or explanation is documented.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- pytorch
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100