Skip to content

Instantly share code, notes, and snippets.

@ilia-cher
Created December 7, 2020 20:28
Show Gist options
  • Save ilia-cher/2a33a207194810c6be73421db8e3f6c6 to your computer and use it in GitHub Desktop.
Save ilia-cher/2a33a207194810c6be73421db8e3f6c6 to your computer and use it in GitHub Desktop.
(pytorch) iliacher@devgpu083:~/local/pytorch (feeee76e)$ ./build/bin/record_function_benchmark
Warmup time: 335 us.
Running without observers
Tensor GEMM benchmark (1x1, 10000): 11665 us.
Tensor GEMM benchmark (16x16, 10000): 52187 us.
Pure RecordFunction benchmark (10000): 155 us.
Running with empty observers
Tensor GEMM benchmark (1x1, 10000): 21440 us.
Tensor GEMM benchmark (16x16, 10000): 61519 us.
Pure RecordFunction benchmark (10000): 1561 us.
Checking number of sampled observer invocations
Pure RecordFunction runtime of 10000000 iterations: 335 us, number of callback invocations: 978, expected number: ~1000 invocations
(pytorch) iliacher@devgpu083:~/local/pytorch (feeee76e)$ ./build/bin/record_function_benchmark
Warmup time: 428 us.
Running without observers
Tensor GEMM benchmark (1x1, 10000): 12193 us.
Tensor GEMM benchmark (16x16, 10000): 51182 us.
Pure RecordFunction benchmark (10000): 190 us.
Running with empty observers
Tensor GEMM benchmark (1x1, 10000): 20991 us.
Tensor GEMM benchmark (16x16, 10000): 58900 us.
Pure RecordFunction benchmark (10000): 1577 us.
Checking number of sampled observer invocations
Pure RecordFunction runtime of 10000000 iterations: 428 us, number of callback invocations: 1012, expected number: ~1000 invocations
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment