microsoft / microsoft/TRELLIS.2
How long where the flow models trained for?
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 11.3k
- Forks
- 1.4k
- PR merge metrics
- No merged PRs in 30d
Description
The provided training configs have max_steps=1,000,000 which seems like a lot. Where the models trained for that long or where they stopped early? I'm trying to replicate the reported results by training TRELLIS.2 from scratch using by own dataset of ~700k meshes. The loss curves seem to be stagnating after training for ~100k iterations but the results aren't yet close to the ones from the released models. Additional details would be very helpful here.
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start with the provided training configs, especially max_steps=1,000,000, and trace the training setup used for the released models. Confirm whether training ran to that limit or stopped early, then document the observed stopping point and relevant details needed to reproduce the reported results.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Documentation
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 45/100