WendaShi

one bash is not correct of quickstart/deploy_to_triton/README.md

4

## Description In Step 1 of quickstart/deploy_to_triton/README.md. , here is a bash: `polygraphy inspect model model.plan --mode=basic` and a error come out as following: `polygraphy inspect model: error: argument --model-type:...

Too many wrong positions are detected

3

When I try to use 'text-editing' mode, as following: `# 2. text editing mode = 'text-editing' input_data = { "prompt": 'The text is centered on a purple background and is...

Thanks for your awesome work❤️ When I use the share captioner (/tools/share-cap_batch_infer.py) to generate captions for some artistic words, I meet some wired captions, but it's not very common. And...

Why didn't covert t5xxl into engine in sd3 pipeline?

3

Thanks for the awesome demo🙏 But when I using the demo by: python3 demo_txt2img_sd3.py "A vibrant street wall covered in colorful graffiti, the centerpiece spells \"SD3 MEDIUM\", in a storm...

If `--no_prior_preservation` is not used, there will be a problem. batch["input_ids"] becomes torch.Size([1, 77]). At this time, in batch["input_ids"][curr_cond_batch_idx], curr_cond_batch_idx=1. The following error occurred: ``` index 1 is out of...

Thanks for this awesome work❤️ It appears that accurately establishing supervision for non-watertight meshes during training is challenging, which may cause the model only to generate watertight meshes. So, is...