alibaba / alibaba/TinyNeuralNetwork
后端可以考虑接入MNN
Open
enhancement
- Dominant language
- Python
- Stars
- 879
- Forks
- 134
- PR merge metrics
- No merged PRs in 30d
Description
现在移动端部署接入了TFLite，需要实现Torch到TFLite的转换；建议后端接入MNN， MNN后端性能领先于TFLite, 且目前提供了torchscript -> mnn模型的转换工具。
Contributor guide
Research direction
No file or test is named in the issue. Start by reviewing the existing TFLite mobile deployment and Torch-to-TFLite conversion paths, then define the MNN backend scope and conversion validation needed for the work to be complete.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- backend-api-design, machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100