RVC-Project / RVC-Project/Retrieval-based-Voice-Conversion-WebUI
models_onnx.py中SineGen.forward()实现为啥这么复杂
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 38.4k
- Forks
- 5.3k
- PR merge metrics
- No merged PRs in 30d
Description
不是只在export时使用吗，f0的batch_size就是1，中间就是时间帧数，最后一维是自己加的，直接生成正弦波就好了，这段代码看着复杂。export()导出时采用动态控制流容易报错。静态图导出的输出结果和torch模型推理有差异。
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
Read models_onnx.py, especially SineGen.forward() and export(), to understand the current tensor shapes and control flow. Compare ONNX static-graph export output with PyTorch inference, then determine whether the implementation can be simplified without changing results or export behavior; done means the discrepancy and export errors are resolved.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- machine-learning
- Issue type
- Refactor
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100