update新结构

This commit is contained in:
2026-01-26 18:27:41 +08:00
parent bc838d7cd7
commit cb610281ce
6 changed files with 55 additions and 2 deletions

View File

@@ -72,3 +72,4 @@ python example/run_pipeline.py --device auto
- The script only samples the first 5000 rows to stay fast.
- `prepare_data.py` runs without PyTorch, but `train.py` and `sample.py` require it.
- `train.py` and `sample.py` auto-select GPU if available; otherwise they fall back to CPU.
- Optional feature-graph mixer (`model_use_feature_graph`) adds a learnable relation prior across feature channels.