简化版的训练框架，全部由C++实现，包含了autograd机制、kernel层关键算子实现，测例中包含单点验证以及端到端训练GPT2模型流程，支持CPU和CUDA平台。
TL17-maker/TinyInfiniTrain
Folders and files
| Name | Name | Last commit date | ||
|---|---|---|---|---|
Repository files navigation
Releases
No releases published
Languages
- C++ 68.6%
- Cuda 28.4%
- Python 1.5%
- Other 1.5%