Accelergy-Project
Question about "Mapspace Dimension [LoopPermutation] Size: 36" in lab of timeloop05
docker% python3 run_example.py 05_null
================================================================================ Calling exercise 05_null. Code is: def run_exercise_05_null(out_dir: str = "output"): start_dir = os.path.join(THIS_SCRIPT_DIR, "05-mapper-conv1d+oc-3level") # Null constraints spec = tl.Specification.from_yaml_files( os.path.join(start_dir, "arch/.yaml"), os.path.join(start_dir, "constraints/null.constraints.yaml"), os.path.join(start_dir, "prob/.yaml"), os.path.join(start_dir, "mapper/*.yaml"), ) tl.call_mapper(spec, output_dir=os.path.join(start_dir, f"{out_dir}/null"))
input file: /home/workspace/tutorial_exercises/01_accelergy_timeloop_2020_ispass/timeloop/05-mapper-conv1d+oc-3level/output/null/parsed-processed-input.yaml
/_ () ___ ___ / /___ ____ ____
/ / / / __ `__ / _ / / __ / __ / __
/ / / / / / / / / / / // / // / // /
// /// // //_//_/_/ ._/
//
Problem configuration complete. execute:/usr/local/bin/accelergy /home/workspace/tutorial_exercises/01_accelergy_timeloop_2020_ispass/timeloop/05-mapper-conv1d+oc-3level/output/null/parsed-processed-input.yaml --oprefix timeloop-mapper. -o ./ > timeloop-mapper.accelergy.log 2>&1 Generate Accelergy ERT (energy reference table) to replace internal energy model. Generate Accelergy ART (area reference table) to replace internal area model. Architecture configuration complete. Sparse optimization configuration complete. Using threads = 1 WARNING: no optimization metric(s) specified, using edp as default. Mapper configuration complete. Initializing Index Factorization subspace. Factorization options along problem dimension K = 21 Factorization options along problem dimension R = 3 Factorization options along problem dimension P = 15 Mapspace Dimension [IndexFactorization] Size: 945 Mapspace Dimension [LoopPermutation] Size: 36
Why is it "36" of size of "Dimension [LoopPermutation]" without constraint but not 6X6X6 ? Is it because of R=3 ? I will very appreciate for your answer. Thank you!
