mirror of
https://github.com/Extraltodeus/test_my_prompt.git
synced 2026-11-05 00:17:26 +00:00
[GH-ISSUE #12] Only output text to grid image? #10
Labels
No labels
pull-request
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
Extraltodeus/test_my_prompt#10
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @ice051128 on GitHub (Dec 2, 2022).
Original GitHub issue: https://github.com/Extraltodeus/test_my_prompt/issues/12
The current implementation renders the text prompts onto all output images then stitch them back into a single grid image, and that makes it impractical and impossible to get back the original output image that u like (especially if the seed is random) since the process is destructive. Even if u use "reuse seed from last generation", the output image is just a slight variation of the original and won't be the same (refer to screenshot to see the issue). And even with a fixed seed, u would then need to run the prompt again without the script, which is inefficient. Is there a way to make it so that it only outputs the text to the grid image, while keeping each individual output image untouched?
Example:
Say u like the picture with no sunset, but the script already renders the text onto that output image so u can't get it back without the text.
Here u can see that the output image is actually different from the original even if it's generated without sunset and with "reuse seed from last generation".
If there's a way to only output text to grid image, it will improve the prompt exploration process and art workflow without any downside since the grid image is what matters for comparison and sometimes there may be stuff that u like during generation.