NVIDIA-Merlin / NVIDIA-Merlin/Merlin
[FEA]Transformer4rec performance enhancement - Try with TF32
Open
Nobody has claimed this yet.
enhancement
- Dominant language
- Python
- Stars
- 907
- Forks
- 129
- PR merge metrics
- No merged PRs in 30d
Description
🚀 Feature request
This is a ticket to enhance t4r performance - one option is to try with lower precision, such as TF32.
Contributor guide
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
The issue names Transformer4rec and TF32 but does not identify files, tests, or a benchmark. Start by locating the Transformer4rec training or inference entry point and existing precision configuration; done should include an agreed TF32 evaluation and evidence of its performance impact.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, tensorflow
- Domain
- machine-learning, performance
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100