HT.Guo

icon indicating a website link hengtaoguo.github.io icon indicating an email [email protected]

icon company Google Cloud icon location Seattle icon location PhD in Biomedical Engineering@RPI Software Engineer@Google Cloud

Results 5 comments of HT.Guo

Convert SMPL-X to SMPL

I tried to make one sample case working by starting with the sample configuration in `smplx2smpl.yaml`. By converting SMPL-X to SMPL, specifically, how should we specify the source model .obj...

Add preprocessing utils for Qwen3-Omni

> is the functionality implemented on cpu in numpy in the torch variant. if so, is there a reason not to want to reuse it? This has been a long-standing...

Review Data Input Pipeline guides

Hi @melissawm could you squash into one commit and sync to head?

Add tokenizer's chat template to SFT

> > > we're modifying the SFT code to also adopt the tokenizer's chat template > > > > > > Could you add the template to the PR description?...

Migrate Gpt3 to NNX.

This PR has been merged by copybara in the following commit, after resolving some g3 conflicts (cc @SurbhiJainUSC ). https://github.com/AI-Hypercomputer/maxtext/commit/c7acd26cf0e37d40982c08aa86f561e4a43c48e1 We will close this one if no issues are detected...