Workaround for SD alt compilation/demo for T4 SM75 #786
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
Currently
compile_alt.py
generates bad images on T4 GPU (SM75).Related issue - #781
Notebook to reproduce the issue with bad images on T4 GPU - AIT_alt_bad_image.ipynb
I found that the issue with bad images can be fixed if we use the following workaround:
I understand that the workaround above uses magic numbers.
But it works...
Testing
Tested on T4 and A100 GPUs .
Compiled with different batch ranges (1-8, 2-4, 1-9) and run demo_alt with different batch sizes (1,2,4,8,9) - all images look ok.