export_aot_inductor() — pytorch Function Reference

Architecture documentation for the export_aot_inductor() function in common.py from the pytorch codebase.

Function python DynamoBenchmarks Configuration calls 2 called by 1

Entity Profile

DynamoBenchmarks→ Configuration→ export_aot_inductor() — pytorch Function Reference

Dependency Diagram

graph TD
  d89eb84e_a6ed_8b3f_56e7_7c49cca94d00["export_aot_inductor()"]
  04a3a4a6_8db3_854d_a893_02c9542bf9dd["speedup_experiment()"]
  04a3a4a6_8db3_854d_a893_02c9542bf9dd -->|calls| d89eb84e_a6ed_8b3f_56e7_7c49cca94d00
  1eff8423_1f23_d138_6815_07d8dc29a749["_normalize_bench_inputs()"]
  d89eb84e_a6ed_8b3f_56e7_7c49cca94d00 -->|calls| 1eff8423_1f23_d138_6815_07d8dc29a749
  3df8de63_d71a_714c_93a1_1a0e24c0f362["load()"]
  d89eb84e_a6ed_8b3f_56e7_7c49cca94d00 -->|calls| 3df8de63_d71a_714c_93a1_1a0e24c0f362
  style d89eb84e_a6ed_8b3f_56e7_7c49cca94d00 fill:#6366f1,stroke:#818cf8,color:#fff

Relationship Graph

Source Code

benchmarks/dynamo/common.py lines 1558–1565

def export_aot_inductor(model, example_inputs, mode):
    optimized = AOTInductorModelCache.load(model, example_inputs, mode)

    def opt_aot_inductor(_, example_inputs, collect_outputs=False):
        example_args, example_kwargs = _normalize_bench_inputs(example_inputs)
        return optimized(*example_args, **example_kwargs)

    return opt_aot_inductor