When will the complete training code be released?

[sx0404]

When will the complete training code be released?

[lawrence-cj]

You may access the whole training code now.

[levi]

Will you also release the LLaVa based captioning pipeline?

[lawrence-cj]

The pipeline is discussed in our paper, and the process is easy and simple. Btw, we will release the LLaVA-captioned SAM dataset in the near future.

[alfredplpl]

Great work! Btw, we need LLaVA-captioned SAM dataset to train model by SA-1B. At the Meta page, the original SA-1B contains only jpg and json. We need the conversion code?

[yujincheng08]

we will release the LLaVA-captioned SAM dataset

[sx0404]

Thank you！！！ I have another greedy request. I would like to ask about the opening hours of weights.

[lawrence-cj]

We will release the caption keeping the original jpg and json name unchanged. On Oct 28, 2023, at 17:45, Alfred Increment ***@***.***> wrote: Great work! Btw, we need LLaVA-captioned SAM dataset to train model by SAM. At the Meta page, it contains only jpg and json. We need the conversion code? —Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you commented.Message ID: ***@***.***>

[alfredplpl]

Thank you so much. I will wait for the release.

[adhikjoshi]

When diffusers pipeline will be available?

[lawrence-cj]

Collaborating with Diffusers & Huggingface team. Plz stay tuned.