get_compile_time() — pytorch Function Reference

Architecture documentation for the get_compile_time() function in cachebench.py from the pytorch codebase.

Function python DynamoBenchmarks Microbenchmarking called by 1

Entity Profile

DynamoBenchmarks→ Microbenchmarking→ get_compile_time() — pytorch Function Reference

Dependency Diagram

graph TD
  a0be0909_592c_396f_91d6_4ab57192f40a["get_compile_time()"]
  0888c17f_2107_6a74_1c6a_de217ecb2d2a["_run_torchbench_from_args()"]
  0888c17f_2107_6a74_1c6a_de217ecb2d2a -->|calls| a0be0909_592c_396f_91d6_4ab57192f40a
  style a0be0909_592c_396f_91d6_4ab57192f40a fill:#6366f1,stroke:#818cf8,color:#fff

Relationship Graph

Source Code

benchmarks/dynamo/cachebench.py lines 46–52

def get_compile_time(file: tempfile._TemporaryFileWrapper) -> float:
    lines = file.readlines()
    # Decode from byte string, remove new lines, parse csv
    lines = [line.decode("utf-8").strip().split(",") for line in lines]
    compilation_time_idx = lines[0].index("compilation_latency")
    compilation_time = lines[1][compilation_time_idx]
    return float(compilation_time)