facebookresearch / facebookresearch/detectron2
Support TensorRT coversion from Detic?
Open
enhancement
- Dominant language
- Python
- Stars
- 34.7k
- Forks
- 7.9k
- PR merge metrics
- No merged PRs in 30d
Description
## 🚀 Feature
Detic model conversion to TensorRT
## Motivation & Examples
Fast inference on production
Contributor guide
Research direction
Start by investigating how Detic models are represented in Detectron2 and whether the repository exposes an existing TensorRT conversion entry point. Compare the current Detic inference path with the production fast-inference goal; done means a documented, working conversion path for Detic models to TensorRT, with validation of inference behavior and performance.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- 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