Reproducibility of results
MrMois opened this issue · comments
Hey, first of all great work!
However, I struggle to reproduce your qualitative visual results from the first page:
results.
I am using classifier free guidance with strength 5. What can I do to further improve the quality?
Best!
Hey there,
the results are cherrypicked and are usually the best out of 10-20 samples. We do hope we get better results with training a bigger model.