main() — pytorch Function Reference

Architecture documentation for the main() function in fx_microbenchmarks.py from the pytorch codebase.

Function python DynamoBenchmarks Microbenchmarking calls 1

Entity Profile

DynamoBenchmarks→ Microbenchmarking→ main() — pytorch Function Reference

Dependency Diagram

graph TD
  6760b199_72fa_746c_6cb0_c904d0485466["main()"]
  b5c9c5fb_ec6e_0bae_bf95_77dadef24243["huge_graph()"]
  6760b199_72fa_746c_6cb0_c904d0485466 -->|calls| b5c9c5fb_ec6e_0bae_bf95_77dadef24243
  style 6760b199_72fa_746c_6cb0_c904d0485466 fill:#6366f1,stroke:#818cf8,color:#fff

Relationship Graph

Source Code

benchmarks/dynamo/microbenchmarks/fx_microbenchmarks.py lines 19–27

def main():
    g = huge_graph()

    def fn():
        for n in g.graph.nodes:
            pass

    t = min(timeit.repeat(fn, number=K, repeat=3))
    print(f"iterating over {N * K} FX nodes took {t:.1f}s ({N * K / t:.0f} nodes/s)")