flexflow / FlexFlow

FlexFlow Serve: Low-Latency, High-Performance LLM Serving
https://flexflow.readthedocs.io
Apache License 2.0
1.6k stars 219 forks source link

Refactor the results of different levels of search #1324

Open wmdi opened 4 months ago

wmdi commented 4 months ago

Refactor the representation of the results of different levels of search, e.g., Strategy. It seems unreasonable to put runtime as part of Strategy since it is determined by pcg and machine_mapping. Also, whether to use visitable for Strategy should be considered.