This commit is contained in:
garrett4wade 2025-07-07 09:38:37 +08:00
parent 078d3e1a44
commit 9a06675ba0
1 changed files with 1 additions and 1 deletions

View File

@ -57,7 +57,7 @@ The implementation layer contains concrete implementations of the API interfaces
- **`fsdp_wrapper.py`**: FSDP-based training engine using PyTorch FSDP2 - **`fsdp_wrapper.py`**: FSDP-based training engine using PyTorch FSDP2
- **`trainer/grpo.py`**: GRPO trainer implementation for reinforcement learning - **`trainer/grpo.py`**: GRPO trainer implementation for reinforcement learning
- **`rollout_controller.py`**: Coordinates rollout data collection across workers - **`rollout_controller.py`**: Coordinates rollout data collection across workers
- **`rlvr/`**: RLVR (RL via Verification and Refinement) collector implementations - **`rlvr/`**: RLVR collector implementations
- **`agentic/`**: Agentic collector implementations (math, code tasks) - **`agentic/`**: Agentic collector implementations (math, code tasks)
#### 3. CLI Layer (`cli/`) #### 3. CLI Layer (`cli/`)